3 perlfunc - Perl builtin functions
7 The functions in this section can serve as terms in an expression.
8 They fall into two major categories: list operators and named unary
9 operators. These differ in their precedence relationship with a
10 following comma. (See the precedence table in L<perlop>.) List
11 operators take more than one argument, while unary operators can never
12 take more than one argument. Thus, a comma terminates the argument of
13 a unary operator, but merely separates the arguments of a list
14 operator. A unary operator generally provides a scalar context to its
15 argument, while a list operator may provide either scalar and list
16 contexts for its arguments. If it does both, the scalar arguments will
17 be first, and the list argument will follow. (Note that there can ever
18 be only one list argument.) For instance, splice() has three scalar
19 arguments followed by a list.
21 In the syntax descriptions that follow, list operators that expect a
22 list (and provide list context for the elements of the list) are shown
23 with LIST as an argument. Such a list may consist of any combination
24 of scalar arguments or list values; the list values will be included
25 in the list as if each individual element were interpolated at that
26 point in the list, forming a longer single-dimensional list value.
27 Elements of the LIST should be separated by commas.
29 Any function in the list below may be used either with or without
30 parentheses around its arguments. (The syntax descriptions omit the
31 parentheses.) If you use the parentheses, the simple (but occasionally
32 surprising) rule is this: It I<LOOKS> like a function, therefore it I<IS> a
33 function, and precedence doesn't matter. Otherwise it's a list
34 operator or unary operator, and precedence does matter. And whitespace
35 between the function and left parenthesis doesn't count--so you need to
38 print 1+2+3; # Prints 6.
39 print(1+2) + 3; # Prints 3.
40 print (1+2)+3; # Also prints 3!
41 print +(1+2)+3; # Prints 6.
42 print ((1+2)+3); # Prints 6.
44 If you run Perl with the B<-w> switch it can warn you about this. For
45 example, the third line above produces:
47 print (...) interpreted as function at - line 1.
48 Useless use of integer addition in void context at - line 1.
50 For functions that can be used in either a scalar or list context,
51 non-abortive failure is generally indicated in a scalar context by
52 returning the undefined value, and in a list context by returning the
55 Remember the following rule:
59 =item I<THERE IS NO GENERAL RULE FOR CONVERTING A LIST INTO A SCALAR!>
63 Each operator and function decides which sort of value it would be most
64 appropriate to return in a scalar context. Some operators return the
65 length of the list that would have been returned in a list context. Some
66 operators return the first value in the list. Some operators return the
67 last value in the list. Some operators return a count of successful
68 operations. In general, they do what you want, unless you want
71 =head2 Perl Functions by Category
73 Here are Perl's functions (including things that look like
74 functions, like some of the keywords and named operators)
75 arranged by category. Some functions appear in more
80 =item Functions for SCALARs or strings
82 chomp, chop, chr, crypt, hex, index, lc, lcfirst, length,
83 oct, ord, pack, q/STRING/, qq/STRING/, reverse, rindex,
84 sprintf, substr, tr///, uc, ucfirst, y///
86 =item Regular expressions and pattern matching
88 m//, pos, quotemeta, s///, split, study
90 =item Numeric functions
92 abs, atan2, cos, exp, hex, int, log, oct, rand, sin, sqrt,
95 =item Functions for real @ARRAYs
97 pop, push, shift, splice, unshift
99 =item Functions for list data
101 grep, join, map, qw/STRING/, reverse, sort, unpack
103 =item Functions for real %HASHes
105 delete, each, exists, keys, values
107 =item Input and output functions
109 binmode, close, closedir, dbmclose, dbmopen, die, eof,
110 fileno, flock, format, getc, print, printf, read, readdir,
111 rewinddir, seek, seekdir, select, syscall, sysread,
112 syswrite, tell, telldir, truncate, warn, write
114 =item Functions for fixed length data or records
116 pack, read, syscall, sysread, syswrite, unpack, vec
118 =item Functions for filehandles, files, or directories
120 I<-X>, chdir, chmod, chown, chroot, fcntl, glob, ioctl, link,
121 lstat, mkdir, open, opendir, readlink, rename, rmdir,
122 stat, symlink, umask, unlink, utime
124 =item Keywords related to the control flow of your perl program
126 caller, continue, die, do, dump, eval, exit, goto, last,
127 next, redo, return, sub, wantarray
129 =item Keywords related to scoping
131 caller, import, local, my, package, use
133 =item Miscellaneous functions
135 defined, dump, eval, formline, local, my, reset, scalar,
138 =item Functions for processes and process groups
140 alarm, exec, fork, getpgrp, getppid, getpriority, kill,
141 pipe, qx/STRING/, setpgrp, setpriority, sleep, system,
144 =item Keywords related to perl modules
146 do, import, no, package, require, use
148 =item Keywords related to classes and object-orientedness
150 bless, dbmclose, dbmopen, package, ref, tie, tied, untie, use
152 =item Low-level socket functions
154 accept, bind, connect, getpeername, getsockname,
155 getsockopt, listen, recv, send, setsockopt, shutdown,
158 =item System V interprocess communication functions
160 msgctl, msgget, msgrcv, msgsnd, semctl, semget, semop,
161 shmctl, shmget, shmread, shmwrite
163 =item Fetching user and group info
165 endgrent, endhostent, endnetent, endpwent, getgrent,
166 getgrgid, getgrnam, getlogin, getpwent, getpwnam,
167 getpwuid, setgrent, setpwent
169 =item Fetching network info
171 endprotoent, endservent, gethostbyaddr, gethostbyname,
172 gethostent, getnetbyaddr, getnetbyname, getnetent,
173 getprotobyname, getprotobynumber, getprotoent,
174 getservbyname, getservbyport, getservent, sethostent,
175 setnetent, setprotoent, setservent
177 =item Time-related functions
179 gmtime, localtime, time, times
181 =item Functions new in perl5
183 abs, bless, chomp, chr, exists, formline, glob, import, lc,
184 lcfirst, map, my, no, prototype, qx, qw, readline, readpipe,
185 ref, sub*, sysopen, tie, tied, uc, ucfirst, untie, use
187 * - C<sub> was a keyword in perl4, but in perl5 it is an
188 operator which can be used in expressions.
190 =item Functions obsoleted in perl5
196 =head2 Alphabetical Listing of Perl Functions
206 A file test, where X is one of the letters listed below. This unary
207 operator takes one argument, either a filename or a filehandle, and
208 tests the associated file to see if something is true about it. If the
209 argument is omitted, tests $_, except for C<-t>, which tests STDIN.
210 Unless otherwise documented, it returns C<1> for TRUE and C<''> for FALSE, or
211 the undefined value if the file doesn't exist. Despite the funny
212 names, precedence is the same as any other named unary operator, and
213 the argument may be parenthesized like any other unary operator. The
214 operator may be any of:
216 -r File is readable by effective uid/gid.
217 -w File is writable by effective uid/gid.
218 -x File is executable by effective uid/gid.
219 -o File is owned by effective uid.
221 -R File is readable by real uid/gid.
222 -W File is writable by real uid/gid.
223 -X File is executable by real uid/gid.
224 -O File is owned by real uid.
227 -z File has zero size.
228 -s File has non-zero size (returns size).
230 -f File is a plain file.
231 -d File is a directory.
232 -l File is a symbolic link.
233 -p File is a named pipe (FIFO).
235 -b File is a block special file.
236 -c File is a character special file.
237 -t Filehandle is opened to a tty.
239 -u File has setuid bit set.
240 -g File has setgid bit set.
241 -k File has sticky bit set.
243 -T File is a text file.
244 -B File is a binary file (opposite of -T).
246 -M Age of file in days when script started.
247 -A Same for access time.
248 -C Same for inode change time.
250 The interpretation of the file permission operators C<-r>, C<-R>, C<-w>,
251 C<-W>, C<-x>, and C<-X> is based solely on the mode of the file and the
252 uids and gids of the user. There may be other reasons you can't actually
253 read, write or execute the file. Also note that, for the superuser,
254 C<-r>, C<-R>, C<-w>, and C<-W> always return 1, and C<-x> and C<-X> return
255 1 if any execute bit is set in the mode. Scripts run by the superuser may
256 thus need to do a stat() to determine the actual mode of the
257 file, or temporarily set the uid to something else.
263 next unless -f $_; # ignore specials
267 Note that C<-s/a/b/> does not do a negated substitution. Saying
268 C<-exp($foo)> still works as expected, however--only single letters
269 following a minus are interpreted as file tests.
271 The C<-T> and C<-B> switches work as follows. The first block or so of the
272 file is examined for odd characters such as strange control codes or
273 characters with the high bit set. If too many odd characters (E<gt>30%)
274 are found, it's a C<-B> file, otherwise it's a C<-T> file. Also, any file
275 containing null in the first block is considered a binary file. If C<-T>
276 or C<-B> is used on a filehandle, the current stdio buffer is examined
277 rather than the first block. Both C<-T> and C<-B> return TRUE on a null
278 file, or a file at EOF when testing a filehandle. Because you have to
279 read a file to do the C<-T> test, on most occasions you want to use a C<-f>
280 against the file first, as in C<next unless -f $file && -T $file>.
282 If any of the file tests (or either the stat() or lstat() operators) are given
283 the special filehandle consisting of a solitary underline, then the stat
284 structure of the previous file test (or stat operator) is used, saving
285 a system call. (This doesn't work with C<-t>, and you need to remember
286 that lstat() and C<-l> will leave values in the stat structure for the
287 symbolic link, not the real file.) Example:
289 print "Can do.\n" if -r $a || -w _ || -x _;
292 print "Readable\n" if -r _;
293 print "Writable\n" if -w _;
294 print "Executable\n" if -x _;
295 print "Setuid\n" if -u _;
296 print "Setgid\n" if -g _;
297 print "Sticky\n" if -k _;
298 print "Text\n" if -T _;
299 print "Binary\n" if -B _;
305 Returns the absolute value of its argument.
306 If VALUE is omitted, uses $_.
308 =item accept NEWSOCKET,GENERICSOCKET
310 Accepts an incoming socket connect, just as the accept(2) system call
311 does. Returns the packed address if it succeeded, FALSE otherwise.
312 See example in L<perlipc/"Sockets: Client/Server Communication">.
318 Arranges to have a SIGALRM delivered to this process after the
319 specified number of seconds have elapsed. If SECONDS is not specified,
320 the value stored in $_ is used. (On some machines,
321 unfortunately, the elapsed time may be up to one second less than you
322 specified because of how seconds are counted.) Only one timer may be
323 counting at once. Each call disables the previous timer, and an
324 argument of 0 may be supplied to cancel the previous timer without
325 starting a new one. The returned value is the amount of time remaining
326 on the previous timer.
328 For delays of finer granularity than one second, you may use Perl's
329 syscall() interface to access setitimer(2) if your system supports it,
330 or else see L</select()> below. It is not advised to intermix alarm()
333 If you want to use alarm() to time out a system call you need to use an
334 eval/die pair. You can't rely on the alarm causing the system call to
335 fail with $! set to EINTR because Perl sets up signal handlers to
336 restart system calls on some systems. Using eval/die always works.
339 local $SIG{ALRM} = sub { die "alarm\n" }; # NB \n required
341 $nread = sysread SOCKET, $buffer, $size;
344 die if $@ && $@ ne "alarm\n"; # propagate errors
354 Returns the arctangent of Y/X in the range -PI to PI.
356 For the tangent operation, you may use the POSIX::tan()
357 function, or use the familiar relation:
359 sub tan { sin($_[0]) / cos($_[0]) }
361 =item bind SOCKET,NAME
363 Binds a network address to a socket, just as the bind system call
364 does. Returns TRUE if it succeeded, FALSE otherwise. NAME should be a
365 packed address of the appropriate type for the socket. See the examples in
366 L<perlipc/"Sockets: Client/Server Communication">.
368 =item binmode FILEHANDLE
370 Arranges for the file to be read or written in "binary" mode in operating
371 systems that distinguish between binary and text files. Files that are
372 not in binary mode have CR LF sequences translated to LF on input and LF
373 translated to CR LF on output. Binmode has no effect under Unix; in DOS
374 and similarly archaic systems, it may be imperative--otherwise your
375 DOS-damaged C library may mangle your file. The key distinction between
376 systems that need binmode and those that don't is their text file
377 formats. Systems like Unix and Plan9 that delimit lines with a single
378 character, and that encode that character in C as '\n', do not need
379 C<binmode>. The rest need it. If FILEHANDLE is an expression, the value
380 is taken as the name of the filehandle.
382 =item bless REF,CLASSNAME
386 This function tells the thingy referenced by REF that it is now
387 an object in the CLASSNAME package--or the current package if no CLASSNAME
388 is specified, which is often the case. It returns the reference for
389 convenience, because a bless() is often the last thing in a constructor.
390 Always use the two-argument version if the function doing the blessing
391 might be inherited by a derived class. See L<perlobj> for more about the
392 blessing (and blessings) of objects.
398 Returns the context of the current subroutine call. In a scalar context,
399 returns the caller's package name if there is a caller, that is, if
400 we're in a subroutine or eval() or require(), and the undefined value
401 otherwise. In a list context, returns
403 ($package, $filename, $line) = caller;
405 With EXPR, it returns some extra information that the debugger uses to
406 print a stack trace. The value of EXPR indicates how many call frames
407 to go back before the current one.
409 ($package, $filename, $line, $subroutine,
410 $hasargs, $wantarray, $evaltext, $is_require) = caller($i);
412 Here $subroutine may be C<"(eval)"> if the frame is not a subroutine
413 call, but C<L<eval>>. In such a case additional elements $evaltext and
414 $is_require are set: $is_require is true if the frame is created by
415 C<L<require>> or C<L<use>> statement, $evaltext contains the text of
416 C<L<eval EXPR>> statement. In particular, for C<L<eval BLOCK>>
417 statement $filename is C<"(eval)">, but $evaltext is undefined. (Note
418 also that C<L<use>> statement creates a C<L<require>> frame inside
419 an C<L<eval EXPR>>) frame.
421 Furthermore, when called from within the DB package, caller returns more
422 detailed information: it sets the list variable @DB::args to be the
423 arguments with which that subroutine was invoked.
427 Changes the working directory to EXPR, if possible. If EXPR is
428 omitted, changes to home directory. Returns TRUE upon success, FALSE
429 otherwise. See example under die().
433 Changes the permissions of a list of files. The first element of the
434 list must be the numerical mode, which should probably be an octal
435 number, and which definitely should I<not> a string of octal digits:
436 C<0644> is okay, C<'0644'> is not. Returns the number of files
437 successfully changed. See also L<oct>, if all you have is a string.
439 $cnt = chmod 0755, 'foo', 'bar';
440 chmod 0755, @executables;
441 $mode = '0644'; chmod $mode, 'foo'; # !!! sets mode to --w----r-T
442 $mode = '0644'; chmod oct($mode), 'foo'; # this is better
443 $mode = 0644; chmod $mode, 'foo'; # this is best
451 This is a slightly safer version of chop (see below). It removes any
452 line ending that corresponds to the current value of C<$/> (also known as
453 $INPUT_RECORD_SEPARATOR in the C<English> module). It returns the total
454 number of characters removed from all its arguments. It's often used to
455 remove the newline from the end of an input record when you're worried
456 that the final record may be missing its newline. When in paragraph mode
457 (C<$/ = "">), it removes all trailing newlines from the string. If
458 VARIABLE is omitted, it chomps $_. Example:
461 chomp; # avoid \n on last field
466 You can actually chomp anything that's an lvalue, including an assignment:
469 chomp($answer = <STDIN>);
471 If you chomp a list, each element is chomped, and the total number of
472 characters removed is returned.
480 Chops off the last character of a string and returns the character
481 chopped. It's used primarily to remove the newline from the end of an
482 input record, but is much more efficient than C<s/\n//> because it neither
483 scans nor copies the string. If VARIABLE is omitted, chops $_.
487 chop; # avoid \n on last field
492 You can actually chop anything that's an lvalue, including an assignment:
495 chop($answer = <STDIN>);
497 If you chop a list, each element is chopped. Only the value of the
498 last chop is returned.
500 Note that chop returns the last character. To return all but the last
501 character, use C<substr($string, 0, -1)>.
505 Changes the owner (and group) of a list of files. The first two
506 elements of the list must be the I<NUMERICAL> uid and gid, in that order.
507 Returns the number of files successfully changed.
509 $cnt = chown $uid, $gid, 'foo', 'bar';
510 chown $uid, $gid, @filenames;
512 Here's an example that looks up non-numeric uids in the passwd file:
515 chop($user = <STDIN>);
517 chop($pattern = <STDIN>);
519 ($login,$pass,$uid,$gid) = getpwnam($user)
520 or die "$user not in passwd file";
522 @ary = <${pattern}>; # expand filenames
523 chown $uid, $gid, @ary;
525 On most systems, you are not allowed to change the ownership of the
526 file unless you're the superuser, although you should be able to change
527 the group to any of your secondary groups. On insecure systems, these
528 restrictions may be relaxed, but this is not a portable assumption.
534 Returns the character represented by that NUMBER in the character set.
535 For example, C<chr(65)> is "A" in ASCII. For the reverse, use L<ord>.
537 If NUMBER is omitted, uses $_.
539 =item chroot FILENAME
543 This function works as the system call by the same name: it makes the
544 named directory the new root directory for all further pathnames that
545 begin with a "/" by your process and all of its children. (It doesn't
546 change your current working directory, which is unaffected.) For security
547 reasons, this call is restricted to the superuser. If FILENAME is
548 omitted, does chroot to $_.
550 =item close FILEHANDLE
552 Closes the file or pipe associated with the file handle, returning TRUE
553 only if stdio successfully flushes buffers and closes the system file
554 descriptor. You don't have to close FILEHANDLE if you are immediately
555 going to do another open() on it, because open() will close it for you. (See
556 open().) However, an explicit close on an input file resets the line
557 counter ($.), while the implicit close done by open() does not. Also,
558 closing a pipe will wait for the process executing on the pipe to
559 complete, in case you want to look at the output of the pipe
560 afterwards. Closing a pipe explicitly also puts the status value of
561 the command into C<$?>. Example:
563 open(OUTPUT, '|sort >foo'); # pipe to sort
564 ... # print stuff to output
565 close OUTPUT; # wait for sort to finish
566 open(INPUT, 'foo'); # get sort's results
568 FILEHANDLE may be an expression whose value gives the real filehandle name.
570 =item closedir DIRHANDLE
572 Closes a directory opened by opendir().
574 =item connect SOCKET,NAME
576 Attempts to connect to a remote socket, just as the connect system call
577 does. Returns TRUE if it succeeded, FALSE otherwise. NAME should be a
578 packed address of the appropriate type for the socket. See the examples in
579 L<perlipc/"Sockets: Client/Server Communication">.
583 Actually a flow control statement rather than a function. If there is a
584 C<continue> BLOCK attached to a BLOCK (typically in a C<while> or
585 C<foreach>), it is always executed just before the conditional is about to
586 be evaluated again, just like the third part of a C<for> loop in C. Thus
587 it can be used to increment a loop variable, even when the loop has been
588 continued via the C<next> statement (which is similar to the C C<continue>
593 Returns the cosine of EXPR (expressed in radians). If EXPR is omitted
596 For the inverse cosine operation, you may use the POSIX::acos()
597 function, or use this relation:
599 sub acos { atan2( sqrt(1 - $_[0] * $_[0]), $_[0] ) }
601 =item crypt PLAINTEXT,SALT
603 Encrypts a string exactly like the crypt(3) function in the C library
604 (assuming that you actually have a version there that has not been
605 extirpated as a potential munition). This can prove useful for checking
606 the password file for lousy passwords, amongst other things. Only the
607 guys wearing white hats should do this.
609 Note that crypt is intended to be a one-way function, much like breaking
610 eggs to make an omelette. There is no (known) corresponding decrypt
611 function. As a result, this function isn't all that useful for
612 cryptography. (For that, see your nearby CPAN mirror.)
614 Here's an example that makes sure that whoever runs this program knows
617 $pwd = (getpwuid($<))[1];
618 $salt = substr($pwd, 0, 2);
622 chop($word = <STDIN>);
626 if (crypt($word, $salt) ne $pwd) {
632 Of course, typing in your own password to whomever asks you
637 [This function has been superseded by the untie() function.]
639 Breaks the binding between a DBM file and a hash.
641 =item dbmopen HASH,DBNAME,MODE
643 [This function has been superseded by the tie() function.]
645 This binds a dbm(3), ndbm(3), sdbm(3), gdbm(), or Berkeley DB file to a
646 hash. HASH is the name of the hash. (Unlike normal open, the first
647 argument is I<NOT> a filehandle, even though it looks like one). DBNAME
648 is the name of the database (without the F<.dir> or F<.pag> extension if
649 any). If the database does not exist, it is created with protection
650 specified by MODE (as modified by the umask()). If your system supports
651 only the older DBM functions, you may perform only one dbmopen() in your
652 program. In older versions of Perl, if your system had neither DBM nor
653 ndbm, calling dbmopen() produced a fatal error; it now falls back to
656 If you don't have write access to the DBM file, you can only read hash
657 variables, not set them. If you want to test whether you can write,
658 either use file tests or try setting a dummy hash entry inside an eval(),
659 which will trap the error.
661 Note that functions such as keys() and values() may return huge array
662 values when used on large DBM files. You may prefer to use the each()
663 function to iterate over large DBM files. Example:
665 # print out history file offsets
666 dbmopen(%HIST,'/usr/lib/news/history',0666);
667 while (($key,$val) = each %HIST) {
668 print $key, ' = ', unpack('L',$val), "\n";
672 See also L<AnyDBM_File> for a more general description of the pros and
673 cons of the various dbm approaches, as well as L<DB_File> for a particularly
680 Returns a Boolean value telling whether EXPR has a value other than
681 the undefined value C<undef>. If EXPR is not present, C<$_> will be
684 Many operations return C<undef> to indicate failure, end of file,
685 system error, uninitialized variable, and other exceptional
686 conditions. This function allows you to distinguish C<undef> from
687 other values. (A simple Boolean test will not distinguish among
688 C<undef>, zero, the empty string, and "0", which are all equally
689 false.) Note that since C<undef> is a valid scalar, its presence
690 doesn't I<necessarily> indicate an exceptional condition: pop()
691 returns C<undef> when its argument is an empty array, I<or> when the
692 element to return happens to be C<undef>.
694 You may also use defined() to check whether a subroutine exists. On
695 the other hand, use of defined() upon aggregates (hashes and arrays)
696 is not guaranteed to produce intuitive results, and should probably be
699 When used on a hash element, it tells you whether the value is defined,
700 not whether the key exists in the hash. Use L<exists> for the latter
705 print if defined $switch{'D'};
706 print "$val\n" while defined($val = pop(@ary));
707 die "Can't readlink $sym: $!"
708 unless defined($value = readlink $sym);
709 sub foo { defined &$bar ? &$bar(@_) : die "No bar"; }
710 $debugging = 0 unless defined $debugging;
712 Note: Many folks tend to overuse defined(), and then are surprised to
713 discover that the number 0 and "" (the zero-length string) are, in fact,
714 defined values. For example, if you say
718 the pattern match succeeds, and $1 is defined, despite the fact that it
719 matched "nothing". But it didn't really match nothing--rather, it
720 matched something that happened to be 0 characters long. This is all
721 very above-board and honest. When a function returns an undefined value,
722 it's an admission that it couldn't give you an honest answer. So you
723 should use defined() only when you're questioning the integrity of what
724 you're trying to do. At other times, a simple comparison to 0 or "" is
727 Currently, using defined() on an entire array or hash reports whether
728 memory for that aggregate has ever been allocated. So an array you set
729 to the empty list appears undefined initially, and one that once was full
730 and that you then set to the empty list still appears defined. You
731 should instead use a simple test for size:
733 if (@an_array) { print "has array elements\n" }
734 if (%a_hash) { print "has hash members\n" }
736 Using undef() on these, however, does clear their memory and then report
737 them as not defined anymore, but you shoudln't do that unless you don't
738 plan to use them again, because it saves time when you load them up
739 again to have memory already ready to be filled.
741 This counter-intuitive behaviour of defined() on aggregates may be
742 changed, fixed, or broken in a future release of Perl.
744 See also L<undef>, L<exists>, L<ref>.
748 Deletes the specified key(s) and their associated values from a hash.
749 For each key, returns the deleted value associated with that key, or
750 the undefined value if there was no such key. Deleting from C<$ENV{}>
751 modifies the environment. Deleting from a hash tied to a DBM file
752 deletes the entry from the DBM file. (But deleting from a tie()d hash
753 doesn't necessarily return anything.)
755 The following deletes all the values of a hash:
757 foreach $key (keys %HASH) {
763 delete @HASH{keys %HASH}
765 (But both of these are slower than the undef() command.) Note that the
766 EXPR can be arbitrarily complicated as long as the final operation is a
767 hash element lookup or hash slice:
769 delete $ref->[$x][$y]{$key};
770 delete @{$ref->[$x][$y]}{$key1, $key2, @morekeys};
774 Outside of an eval(), prints the value of LIST to C<STDERR> and exits with
775 the current value of C<$!> (errno). If C<$!> is 0, exits with the value of
776 C<($? E<gt>E<gt> 8)> (back-tick `command` status). If C<($? E<gt>E<gt> 8)>
777 is 0, exits with 255. Inside an eval(), the error message is stuffed into
778 C<$@>, and the eval() is terminated with the undefined value; this makes
779 die() the way to raise an exception.
783 die "Can't cd to spool: $!\n" unless chdir '/usr/spool/news';
784 chdir '/usr/spool/news' or die "Can't cd to spool: $!\n"
786 If the value of EXPR does not end in a newline, the current script line
787 number and input line number (if any) are also printed, and a newline
788 is supplied. Hint: sometimes appending ", stopped" to your message
789 will cause it to make better sense when the string "at foo line 123" is
790 appended. Suppose you are running script "canasta".
792 die "/etc/games is no good";
793 die "/etc/games is no good, stopped";
795 produce, respectively
797 /etc/games is no good at canasta line 123.
798 /etc/games is no good, stopped at canasta line 123.
800 See also exit() and warn().
802 You can arrange for a callback to be called just before the die() does
803 its deed, by setting the C<$SIG{__DIE__}> hook. The associated handler
804 will be called with the error text and can change the error message, if
805 it sees fit, by calling die() again. See L<perlvar> for details on
806 setting C<%SIG> entries, and eval() for some examples.
810 Not really a function. Returns the value of the last command in the
811 sequence of commands indicated by BLOCK. When modified by a loop
812 modifier, executes the BLOCK once before testing the loop condition.
813 (On other statements the loop modifiers test the conditional first.)
815 =item do SUBROUTINE(LIST)
817 A deprecated form of subroutine call. See L<perlsub>.
821 Uses the value of EXPR as a filename and executes the contents of the
822 file as a Perl script. Its primary use is to include subroutines
823 from a Perl subroutine library.
831 except that it's more efficient, more concise, keeps track of the
832 current filename for error messages, and searches all the B<-I>
833 libraries if the file isn't in the current directory (see also the @INC
834 array in L<perlvar/Predefined Names>). It's the same, however, in that it does
835 re-parse the file every time you call it, so you probably don't want to
836 do this inside a loop.
838 Note that inclusion of library modules is better done with the
839 use() and require() operators, which also do error checking
840 and raise an exception if there's a problem.
844 This causes an immediate core dump. Primarily this is so that you can
845 use the B<undump> program to turn your core dump into an executable binary
846 after having initialized all your variables at the beginning of the
847 program. When the new binary is executed it will begin by executing a
848 C<goto LABEL> (with all the restrictions that C<goto> suffers). Think of
849 it as a goto with an intervening core dump and reincarnation. If LABEL
850 is omitted, restarts the program from the top. WARNING: any files
851 opened at the time of the dump will NOT be open any more when the
852 program is reincarnated, with possible resulting confusion on the part
853 of Perl. See also B<-u> option in L<perlrun>.
870 dump QUICKSTART if $ARGV[0] eq '-d';
877 When called in a list context, returns a 2-element array consisting of the
878 key and value for the next element of a hash, so that you can iterate over
879 it. When called in a scalar context, returns the key for only the next
880 element in the hash. (Note: Keys may be "0" or "", which are logically
881 false; you may wish to avoid constructs like C<while ($k = each %foo) {}>
884 Entries are returned in an apparently random order. When the hash is
885 entirely read, a null array is returned in list context (which when
886 assigned produces a FALSE (0) value), and C<undef> is returned in a
887 scalar context. The next call to each() after that will start iterating
888 again. There is a single iterator for each hash, shared by all each(),
889 keys(), and values() function calls in the program; it can be reset by
890 reading all the elements from the hash, or by evaluating C<keys HASH> or
891 C<values HASH>. If you add or delete elements of a hash while you're
892 iterating over it, you may get entries skipped or duplicated, so don't.
894 The following prints out your environment like the printenv(1) program,
895 only in a different order:
897 while (($key,$value) = each %ENV) {
898 print "$key=$value\n";
901 See also keys() and values().
909 Returns 1 if the next read on FILEHANDLE will return end of file, or if
910 FILEHANDLE is not open. FILEHANDLE may be an expression whose value
911 gives the real filehandle name. (Note that this function actually
912 reads a character and then ungetc()s it, so it is not very useful in an
913 interactive context.) Do not read from a terminal file (or call
914 C<eof(FILEHANDLE)> on it) after end-of-file is reached. Filetypes such
915 as terminals may lose the end-of-file condition if you do.
917 An C<eof> without an argument uses the last file read as argument.
918 Empty parentheses () may be used to indicate the pseudo file formed of
919 the files listed on the command line, i.e., C<eof()> is reasonable to
920 use inside a C<while (E<lt>E<gt>)> loop to detect the end of only the
921 last file. Use C<eof(ARGV)> or eof without the parentheses to test
922 I<EACH> file in a while (E<lt>E<gt>) loop. Examples:
924 # reset line numbering on each input file
927 close(ARGV) if (eof); # Not eof().
930 # insert dashes just before last line of last file
933 print "--------------\n";
934 close(ARGV); # close or break; is needed if we
935 # are reading from the terminal
940 Practical hint: you almost never need to use C<eof> in Perl, because the
941 input operators return undef when they run out of data.
947 EXPR is parsed and executed as if it were a little Perl program. It
948 is executed in the context of the current Perl program, so that any
949 variable settings or subroutine and format definitions remain afterwards.
950 The value returned is the value of the last expression evaluated, or a
951 return statement may be used, just as with subroutines. The last
952 expression is evaluated in scalar or array context, depending on the
955 If there is a syntax error or runtime error, or a die() statement is
956 executed, an undefined value is returned by eval(), and C<$@> is set to the
957 error message. If there was no error, C<$@> is guaranteed to be a null
958 string. If EXPR is omitted, evaluates C<$_>. The final semicolon, if
959 any, may be omitted from the expression. Beware that using eval()
960 neither silences perl from printing warnings to STDERR, nor does it
961 stuff the text of warning messages into C<$@>. To do either of those,
962 you have to use the C<$SIG{__WARN__}> facility. See warn() and L<perlvar>.
964 Note that, because eval() traps otherwise-fatal errors, it is useful for
965 determining whether a particular feature (such as socket() or symlink())
966 is implemented. It is also Perl's exception trapping mechanism, where
967 the die operator is used to raise exceptions.
969 If the code to be executed doesn't vary, you may use the eval-BLOCK
970 form to trap run-time errors without incurring the penalty of
971 recompiling each time. The error, if any, is still returned in C<$@>.
974 # make divide-by-zero non-fatal
975 eval { $answer = $a / $b; }; warn $@ if $@;
977 # same thing, but less efficient
978 eval '$answer = $a / $b'; warn $@ if $@;
980 # a compile-time error
984 eval '$answer ='; # sets $@
986 When using the eval{} form as an exception trap in libraries, you may
987 wish not to trigger any C<__DIE__> hooks that user code may have
988 installed. You can use the C<local $SIG{__DIE__}> construct for this
989 purpose, as shown in this example:
991 # a very private exception trap for divide-by-zero
992 eval { local $SIG{'__DIE__'}; $answer = $a / $b; }; warn $@ if $@;
994 This is especially significant, given that C<__DIE__> hooks can call
995 die() again, which has the effect of changing their error messages:
997 # __DIE__ hooks may modify error messages
999 local $SIG{'__DIE__'} = sub { (my $x = $_[0]) =~ s/foo/bar/g; die $x };
1000 eval { die "foo foofs here" };
1001 print $@ if $@; # prints "bar barfs here"
1004 With an eval(), you should be especially careful to remember what's
1005 being looked at when:
1011 eval { $x }; # CASE 4
1013 eval "\$$x++" # CASE 5
1016 Cases 1 and 2 above behave identically: they run the code contained in
1017 the variable $x. (Although case 2 has misleading double quotes making
1018 the reader wonder what else might be happening (nothing is).) Cases 3
1019 and 4 likewise behave in the same way: they run the code '$x', which
1020 does nothing but return the value of C<$x>. (Case 4 is preferred for
1021 purely visual reasons, but it also has the advantage of compiling at
1022 compile-time instead of at run-time.) Case 5 is a place where
1023 normally you I<WOULD> like to use double quotes, except that in that
1024 particular situation, you can just use symbolic references instead, as
1029 The exec() function executes a system command I<AND NEVER RETURNS>,
1030 unless the command does not exist and is executed directly instead of
1031 via C</bin/sh -c> (see below). Use system() instead of exec() if you
1034 If there is more than one argument in LIST, or if LIST is an array with
1035 more than one value, calls execvp(3) with the arguments in LIST. If
1036 there is only one scalar argument, the argument is checked for shell
1037 metacharacters. If there are any, the entire argument is passed to
1038 C</bin/sh -c> for parsing. If there are none, the argument is split
1039 into words and passed directly to execvp(), which is more efficient.
1040 Note: exec() and system() do not flush your output buffer, so you may
1041 need to set C<$|> to avoid lost output. Examples:
1043 exec '/bin/echo', 'Your arguments are: ', @ARGV;
1044 exec "sort $outfile | uniq";
1046 If you don't really want to execute the first argument, but want to lie
1047 to the program you are executing about its own name, you can specify
1048 the program you actually want to run as an "indirect object" (without a
1049 comma) in front of the LIST. (This always forces interpretation of the
1050 LIST as a multi-valued list, even if there is only a single scalar in
1053 $shell = '/bin/csh';
1054 exec $shell '-sh'; # pretend it's a login shell
1058 exec {'/bin/csh'} '-sh'; # pretend it's a login shell
1062 Returns TRUE if the specified hash key exists in its hash array, even
1063 if the corresponding value is undefined.
1065 print "Exists\n" if exists $array{$key};
1066 print "Defined\n" if defined $array{$key};
1067 print "True\n" if $array{$key};
1069 A hash element can be TRUE only if it's defined, and defined if
1070 it exists, but the reverse doesn't necessarily hold true.
1072 Note that the EXPR can be arbitrarily complicated as long as the final
1073 operation is a hash key lookup:
1075 if (exists $ref->[$x][$y]{$key}) { ... }
1079 Evaluates EXPR and exits immediately with that value. (Actually, it
1080 calls any defined C<END> routines first, but the C<END> routines may not
1081 abort the exit. Likewise any object destructors that need to be called
1082 are called before exit.) Example:
1085 exit 0 if $ans =~ /^[Xx]/;
1087 See also die(). If EXPR is omitted, exits with 0 status. The only
1088 univerally portable values for EXPR are 0 for success and 1 for error;
1089 all other values are subject to unpredictable interpretation depending
1090 on the environment in which the Perl program is running.
1092 You shouldn't use exit() to abort a subroutine if there's any chance that
1093 someone might want to trap whatever error happened. Use die() instead,
1094 which can be trapped by an eval().
1100 Returns I<e> (the natural logarithm base) to the power of EXPR.
1101 If EXPR is omitted, gives C<exp($_)>.
1103 =item fcntl FILEHANDLE,FUNCTION,SCALAR
1105 Implements the fcntl(2) function. You'll probably have to say
1109 first to get the correct function definitions. Argument processing and
1110 value return works just like ioctl() below. Note that fcntl() will produce
1111 a fatal error if used on a machine that doesn't implement fcntl(2).
1115 fcntl($filehandle, F_GETLK, $packed_return_buffer);
1117 =item fileno FILEHANDLE
1119 Returns the file descriptor for a filehandle. This is useful for
1120 constructing bitmaps for select(). If FILEHANDLE is an expression, the
1121 value is taken as the name of the filehandle.
1123 =item flock FILEHANDLE,OPERATION
1125 Calls flock(2), or an emulation of it, on FILEHANDLE. Returns TRUE for
1126 success, FALSE on failure. Will produce a fatal error if used on a
1127 machine that doesn't implement flock(2), fcntl(2) locking, or lockf(3).
1128 flock() is Perl's portable file locking interface, although it will lock
1129 only entire files, not records.
1131 OPERATION is one of LOCK_SH, LOCK_EX, or LOCK_UN, possibly combined with
1132 LOCK_NB. These constants are traditionally valued 1, 2, 8 and 4, but
1133 you can use the symbolic names if you pull them in with an explicit
1134 request to the Fcntl module. The names can be requested as a group with
1135 the :flock tag (or they can be requested individually, of course).
1136 LOCK_SH requests a shared lock, LOCK_EX requests an exclusive lock, and
1137 LOCK_UN releases a previously requested lock. If LOCK_NB is added to
1138 LOCK_SH or LOCK_EX then flock() will return immediately rather than
1139 blocking waiting for the lock (check the return status to see if you got
1142 Note that the emulation built with lockf(3) doesn't provide shared
1143 locks, and it requires that FILEHANDLE be open with write intent. These
1144 are the semantics that lockf(3) implements. Most (all?) systems
1145 implement lockf(3) in terms of fcntl(2) locking, though, so the
1146 differing semantics shouldn't bite too many people.
1148 Note also that some versions of flock() cannot lock things over the
1149 network; you would need to use the more system-specific fcntl() for
1150 that. If you like you can force Perl to ignore your system's flock(2)
1151 function, and so provide its own fcntl(2)-based emulation, by passing
1152 the switch C<-Ud_flock> to the F<Configure> program when you configure
1155 Here's a mailbox appender for BSD systems.
1157 use Fcntl ':flock'; # import LOCK_* constants
1160 flock(MBOX,LOCK_EX);
1161 # and, in case someone appended
1162 # while we were waiting...
1167 flock(MBOX,LOCK_UN);
1170 open(MBOX, ">>/usr/spool/mail/$ENV{'USER'}")
1171 or die "Can't open mailbox: $!";
1174 print MBOX $msg,"\n\n";
1177 See also L<DB_File> for other flock() examples.
1181 Does a fork(2) system call. Returns the child pid to the parent process
1182 and 0 to the child process, or C<undef> if the fork is unsuccessful.
1183 Note: unflushed buffers remain unflushed in both processes, which means
1184 you may need to set C<$|> ($AUTOFLUSH in English) or call the autoflush()
1185 method of IO::Handle to avoid duplicate output.
1187 If you fork() without ever waiting on your children, you will accumulate
1190 $SIG{CHLD} = sub { wait };
1192 There's also the double-fork trick (error checking on
1193 fork() returns omitted);
1195 unless ($pid = fork) {
1197 exec "what you really wanna do";
1200 ## (some_perl_code_here)
1207 See also L<perlipc> for more examples of forking and reaping
1210 Note that if your forked child inherits system file descriptors like
1211 STDIN and STDOUT that are actually connected by a pipe or socket, even
1212 if you exit, the remote server (such as, say, httpd or rsh) won't think
1213 you're done. You should reopen those to /dev/null if it's any issue.
1217 Declare a picture format with use by the write() function. For
1221 Test: @<<<<<<<< @||||| @>>>>>
1222 $str, $%, '$' . int($num)
1226 $num = $cost/$quantity;
1230 See L<perlform> for many details and examples.
1233 =item formline PICTURE, LIST
1235 This is an internal function used by C<format>s, though you may call it
1236 too. It formats (see L<perlform>) a list of values according to the
1237 contents of PICTURE, placing the output into the format output
1238 accumulator, C<$^A> (or $ACCUMULATOR in English).
1239 Eventually, when a write() is done, the contents of
1240 C<$^A> are written to some filehandle, but you could also read C<$^A>
1241 yourself and then set C<$^A> back to "". Note that a format typically
1242 does one formline() per line of form, but the formline() function itself
1243 doesn't care how many newlines are embedded in the PICTURE. This means
1244 that the C<~> and C<~~> tokens will treat the entire PICTURE as a single line.
1245 You may therefore need to use multiple formlines to implement a single
1246 record format, just like the format compiler.
1248 Be careful if you put double quotes around the picture, because an "C<@>"
1249 character may be taken to mean the beginning of an array name.
1250 formline() always returns TRUE. See L<perlform> for other examples.
1252 =item getc FILEHANDLE
1256 Returns the next character from the input file attached to FILEHANDLE,
1257 or a null string at end of file. If FILEHANDLE is omitted, reads from STDIN.
1258 This is not particularly efficient. It cannot be used to get unbuffered
1259 single-characters, however. For that, try something more like:
1262 system "stty cbreak </dev/tty >/dev/tty 2>&1";
1265 system "stty", '-icanon', 'eol', "\001";
1271 system "stty -cbreak </dev/tty >/dev/tty 2>&1";
1274 system "stty", 'icanon', 'eol', '^@'; # ASCII null
1278 Determination of whether $BSD_STYLE should be set
1279 is left as an exercise to the reader.
1281 The POSIX::getattr() function can do this more portably on systems
1282 alleging POSIX compliance.
1283 See also the C<Term::ReadKey> module from your nearest CPAN site;
1284 details on CPAN can be found on L<perlmod/CPAN>.
1288 Returns the current login from F</etc/utmp>, if any. If null, use
1291 $login = getlogin || getpwuid($<) || "Kilroy";
1293 Do not consider getlogin() for authentication: it is not as
1294 secure as getpwuid().
1296 =item getpeername SOCKET
1298 Returns the packed sockaddr address of other end of the SOCKET connection.
1301 $hersockaddr = getpeername(SOCK);
1302 ($port, $iaddr) = unpack_sockaddr_in($hersockaddr);
1303 $herhostname = gethostbyaddr($iaddr, AF_INET);
1304 $herstraddr = inet_ntoa($iaddr);
1308 Returns the current process group for the specified PID. Use
1309 a PID of 0 to get the current process group for the
1310 current process. Will raise an exception if used on a machine that
1311 doesn't implement getpgrp(2). If PID is omitted, returns process
1312 group of current process. Note that the POSIX version of getpgrp()
1313 does not accept a PID argument, so only PID==0 is truly portable.
1317 Returns the process id of the parent process.
1319 =item getpriority WHICH,WHO
1321 Returns the current priority for a process, a process group, or a user.
1322 (See L<getpriority(2)>.) Will raise a fatal exception if used on a
1323 machine that doesn't implement getpriority(2).
1329 =item gethostbyname NAME
1331 =item getnetbyname NAME
1333 =item getprotobyname NAME
1339 =item getservbyname NAME,PROTO
1341 =item gethostbyaddr ADDR,ADDRTYPE
1343 =item getnetbyaddr ADDR,ADDRTYPE
1345 =item getprotobynumber NUMBER
1347 =item getservbyport PORT,PROTO
1365 =item sethostent STAYOPEN
1367 =item setnetent STAYOPEN
1369 =item setprotoent STAYOPEN
1371 =item setservent STAYOPEN
1385 These routines perform the same functions as their counterparts in the
1386 system library. Within a list context, the return values from the
1387 various get routines are as follows:
1389 ($name,$passwd,$uid,$gid,
1390 $quota,$comment,$gcos,$dir,$shell) = getpw*
1391 ($name,$passwd,$gid,$members) = getgr*
1392 ($name,$aliases,$addrtype,$length,@addrs) = gethost*
1393 ($name,$aliases,$addrtype,$net) = getnet*
1394 ($name,$aliases,$proto) = getproto*
1395 ($name,$aliases,$port,$proto) = getserv*
1397 (If the entry doesn't exist you get a null list.)
1399 Within a scalar context, you get the name, unless the function was a
1400 lookup by name, in which case you get the other thing, whatever it is.
1401 (If the entry doesn't exist you get the undefined value.) For example:
1411 The $members value returned by I<getgr*()> is a space separated list of
1412 the login names of the members of the group.
1414 For the I<gethost*()> functions, if the C<h_errno> variable is supported in
1415 C, it will be returned to you via C<$?> if the function call fails. The
1416 @addrs value returned by a successful call is a list of the raw
1417 addresses returned by the corresponding system library call. In the
1418 Internet domain, each address is four bytes long and you can unpack it
1419 by saying something like:
1421 ($a,$b,$c,$d) = unpack('C4',$addr[0]);
1423 =item getsockname SOCKET
1425 Returns the packed sockaddr address of this end of the SOCKET connection.
1428 $mysockaddr = getsockname(SOCK);
1429 ($port, $myaddr) = unpack_sockaddr_in($mysockaddr);
1431 =item getsockopt SOCKET,LEVEL,OPTNAME
1433 Returns the socket option requested, or undefined if there is an error.
1439 Returns the value of EXPR with filename expansions such as a shell
1440 would do. This is the internal function implementing the
1441 C<E<lt>*.cE<gt>> operator, except it's easier to use. If EXPR is
1442 omitted, $_ is used. The E<lt>E<gt> operator is discussed in more
1443 detail in L<perlop/"I/O Operators">.
1447 Converts a time as returned by the time function to a 9-element array
1448 with the time localized for the standard Greenwich time zone.
1449 Typically used as follows:
1452 ($sec,$min,$hour,$mday,$mon,$year,$wday,$yday,$isdst) =
1455 All array elements are numeric, and come straight out of a struct tm.
1456 In particular this means that $mon has the range 0..11 and $wday has
1457 the range 0..6. Also, $year is the number of years since 1900, I<not>
1458 simply the last two digits of the year.
1460 If EXPR is omitted, does C<gmtime(time())>.
1462 In a scalar context, prints out the ctime(3) value:
1464 $now_string = gmtime; # e.g., "Thu Oct 13 04:54:34 1994"
1466 Also see the F<timegm.pl> library, and the strftime(3) function available
1467 via the POSIX module.
1475 The goto-LABEL form finds the statement labeled with LABEL and resumes
1476 execution there. It may not be used to go into any construct that
1477 requires initialization, such as a subroutine or a foreach loop. It
1478 also can't be used to go into a construct that is optimized away,
1479 or to get out of a block or subroutine given to sort().
1480 It can be used to go almost anywhere else within the dynamic scope,
1481 including out of subroutines, but it's usually better to use some other
1482 construct such as last or die. The author of Perl has never felt the
1483 need to use this form of goto (in Perl, that is--C is another matter).
1485 The goto-EXPR form expects a label name, whose scope will be resolved
1486 dynamically. This allows for computed gotos per FORTRAN, but isn't
1487 necessarily recommended if you're optimizing for maintainability:
1489 goto ("FOO", "BAR", "GLARCH")[$i];
1491 The goto-&NAME form is highly magical, and substitutes a call to the
1492 named subroutine for the currently running subroutine. This is used by
1493 AUTOLOAD subroutines that wish to load another subroutine and then
1494 pretend that the other subroutine had been called in the first place
1495 (except that any modifications to @_ in the current subroutine are
1496 propagated to the other subroutine.) After the goto, not even caller()
1497 will be able to tell that this routine was called first.
1499 =item grep BLOCK LIST
1501 =item grep EXPR,LIST
1503 This is similar in spirit to, but not the same as, L<grep(1)>
1504 and its relatives. In particular, it is not limited to using
1505 regular expressions.
1507 Evaluates the BLOCK or EXPR for each element of LIST (locally setting
1508 $_ to each element) and returns the list value consisting of those
1509 elements for which the expression evaluated to TRUE. In a scalar
1510 context, returns the number of times the expression was TRUE.
1512 @foo = grep(!/^#/, @bar); # weed out comments
1516 @foo = grep {!/^#/} @bar; # weed out comments
1518 Note that, because $_ is a reference into the list value, it can be used
1519 to modify the elements of the array. While this is useful and
1520 supported, it can cause bizarre results if the LIST is not a named
1521 array. Similarly, grep returns aliases into the original list,
1522 much like the way that L<Foreach Loops>'s index variable aliases the list
1523 elements. That is, modifying an element of a list returned by grep
1524 actually modifies the element in the original list.
1530 Interprets EXPR as a hex string and returns the corresponding
1531 value. (To convert strings that might start with either 0 or 0x
1532 see L<oct>.) If EXPR is omitted, uses $_.
1534 print hex '0xAf'; # prints '175'
1535 print hex 'aF'; # same
1539 There is no built-in import() function. It is merely an ordinary
1540 method (subroutine) defined (or inherited) by modules that wish to export
1541 names to another module. The use() function calls the import() method
1542 for the package used. See also L</use>, L<perlmod>, and L<Exporter>.
1544 =item index STR,SUBSTR,POSITION
1546 =item index STR,SUBSTR
1548 Returns the position of the first occurrence of SUBSTR in STR at or after
1549 POSITION. If POSITION is omitted, starts searching from the beginning of
1550 the string. The return value is based at 0 (or whatever you've set the C<$[>
1551 variable to--but don't do that). If the substring is not found, returns
1552 one less than the base, ordinarily -1.
1558 Returns the integer portion of EXPR. If EXPR is omitted, uses $_.
1560 =item ioctl FILEHANDLE,FUNCTION,SCALAR
1562 Implements the ioctl(2) function. You'll probably have to say
1564 require "ioctl.ph"; # probably in /usr/local/lib/perl/ioctl.ph
1566 first to get the correct function definitions. If F<ioctl.ph> doesn't
1567 exist or doesn't have the correct definitions you'll have to roll your
1568 own, based on your C header files such as F<E<lt>sys/ioctl.hE<gt>>.
1569 (There is a Perl script called B<h2ph> that comes with the Perl kit which
1570 may help you in this, but it's non-trivial.) SCALAR will be read and/or
1571 written depending on the FUNCTION--a pointer to the string value of SCALAR
1572 will be passed as the third argument of the actual ioctl call. (If SCALAR
1573 has no string value but does have a numeric value, that value will be
1574 passed rather than a pointer to the string value. To guarantee this to be
1575 TRUE, add a 0 to the scalar before using it.) The pack() and unpack()
1576 functions are useful for manipulating the values of structures used by
1577 ioctl(). The following example sets the erase character to DEL.
1581 die "NO TIOCGETP" if $@ || !$getp;
1582 $sgttyb_t = "ccccs"; # 4 chars and a short
1583 if (ioctl(STDIN,$getp,$sgttyb)) {
1584 @ary = unpack($sgttyb_t,$sgttyb);
1586 $sgttyb = pack($sgttyb_t,@ary);
1587 ioctl(STDIN,&TIOCSETP,$sgttyb)
1588 || die "Can't ioctl: $!";
1591 The return value of ioctl (and fcntl) is as follows:
1593 if OS returns: then Perl returns:
1595 0 string "0 but true"
1596 anything else that number
1598 Thus Perl returns TRUE on success and FALSE on failure, yet you can
1599 still easily determine the actual value returned by the operating
1602 ($retval = ioctl(...)) || ($retval = -1);
1603 printf "System returned %d\n", $retval;
1605 =item join EXPR,LIST
1607 Joins the separate strings of LIST or ARRAY into a single string with
1608 fields separated by the value of EXPR, and returns the string.
1611 $_ = join(':', $login,$passwd,$uid,$gid,$gcos,$home,$shell);
1613 See L<perlfunc/split>.
1617 Returns a normal array consisting of all the keys of the named hash. (In
1618 a scalar context, returns the number of keys.) The keys are returned in
1619 an apparently random order, but it is the same order as either the
1620 values() or each() function produces (given that the hash has not been
1621 modified). As a side effect, it resets HASH's iterator.
1623 Here is yet another way to print your environment:
1626 @values = values %ENV;
1627 while ($#keys >= 0) {
1628 print pop(@keys), '=', pop(@values), "\n";
1631 or how about sorted by key:
1633 foreach $key (sort(keys %ENV)) {
1634 print $key, '=', $ENV{$key}, "\n";
1637 To sort an array by value, you'll need to use a C<sort{}> function.
1638 Here's a descending numeric sort of a hash by its values:
1640 foreach $key (sort { $hash{$b} <=> $hash{$a} } keys %hash)) {
1641 printf "%4d %s\n", $hash{$key}, $key;
1644 As an lvalue C<keys> allows you to increase the number of hash buckets
1645 allocated for the given hash. This can gain you a measure of efficiency if
1646 you know the hash is going to get big. (This is similar to pre-extending
1647 an array by assigning a larger number to $#array.) If you say
1651 then C<%hash> will have at least 200 buckets allocated for it. These
1652 buckets will be retained even if you do C<%hash = ()>, use C<undef
1653 %hash> if you want to free the storage while C<%hash> is still in scope.
1654 You can't shrink the number of buckets allocated for the hash using
1655 C<keys> in this way (but you needn't worry about doing this by accident,
1656 as trying has no effect).
1660 Sends a signal to a list of processes. The first element of
1661 the list must be the signal to send. Returns the number of
1662 processes successfully signaled.
1664 $cnt = kill 1, $child1, $child2;
1667 Unlike in the shell, in Perl if the I<SIGNAL> is negative, it kills
1668 process groups instead of processes. (On System V, a negative I<PROCESS>
1669 number will also kill process groups, but that's not portable.) That
1670 means you usually want to use positive not negative signals. You may also
1671 use a signal name in quotes. See L<perlipc/"Signals"> for details.
1677 The C<last> command is like the C<break> statement in C (as used in
1678 loops); it immediately exits the loop in question. If the LABEL is
1679 omitted, the command refers to the innermost enclosing loop. The
1680 C<continue> block, if any, is not executed:
1682 LINE: while (<STDIN>) {
1683 last LINE if /^$/; # exit when done with header
1691 Returns an lowercased version of EXPR. This is the internal function
1692 implementing the \L escape in double-quoted strings.
1693 Respects current LC_CTYPE locale if C<use locale> in force. See L<perllocale>.
1695 If EXPR is omitted, uses $_.
1701 Returns the value of EXPR with the first character lowercased. This is
1702 the internal function implementing the \l escape in double-quoted strings.
1703 Respects current LC_CTYPE locale if C<use locale> in force. See L<perllocale>.
1705 If EXPR is omitted, uses $_.
1711 Returns the length in characters of the value of EXPR. If EXPR is
1712 omitted, returns length of $_.
1714 =item link OLDFILE,NEWFILE
1716 Creates a new filename linked to the old filename. Returns 1 for
1717 success, 0 otherwise.
1719 =item listen SOCKET,QUEUESIZE
1721 Does the same thing that the listen system call does. Returns TRUE if
1722 it succeeded, FALSE otherwise. See example in L<perlipc/"Sockets: Client/Server Communication">.
1726 A local modifies the listed variables to be local to the enclosing block,
1727 subroutine, C<eval{}>, or C<do>. If more than one value is listed, the
1728 list must be placed in parentheses. See L<perlsub/"Temporary Values via
1729 local()"> for details.
1731 But you really probably want to be using my() instead, because local() isn't
1732 what most people think of as "local"). See L<perlsub/"Private Variables
1733 via my()"> for details.
1735 =item localtime EXPR
1737 Converts a time as returned by the time function to a 9-element array
1738 with the time analyzed for the local time zone. Typically used as
1741 ($sec,$min,$hour,$mday,$mon,$year,$wday,$yday,$isdst) =
1744 All array elements are numeric, and come straight out of a struct tm.
1745 In particular this means that $mon has the range 0..11 and $wday has
1746 the range 0..6 and $year is year-1900, that is, $year is 123 in year
1747 2023. If EXPR is omitted, uses the current time ("localtime(time)").
1749 In a scalar context, returns the ctime(3) value:
1751 $now_string = localtime; # e.g., "Thu Oct 13 04:54:34 1994"
1753 Also see the Time::Local module, and the strftime(3) function available
1754 via the POSIX module.
1760 Returns logarithm (base I<e>) of EXPR. If EXPR is omitted, returns log
1763 =item lstat FILEHANDLE
1769 Does the same thing as the stat() function, but stats a symbolic link
1770 instead of the file the symbolic link points to. If symbolic links are
1771 unimplemented on your system, a normal stat() is done.
1773 If EXPR is omitted, stats $_.
1777 The match operator. See L<perlop>.
1779 =item map BLOCK LIST
1783 Evaluates the BLOCK or EXPR for each element of LIST (locally setting $_ to each
1784 element) and returns the list value composed of the results of each such
1785 evaluation. Evaluates BLOCK or EXPR in a list context, so each element of LIST
1786 may produce zero, one, or more elements in the returned value.
1788 @chars = map(chr, @nums);
1790 translates a list of numbers to the corresponding characters. And
1792 %hash = map { getkey($_) => $_ } @array;
1794 is just a funny way to write
1797 foreach $_ (@array) {
1798 $hash{getkey($_)} = $_;
1801 =item mkdir FILENAME,MODE
1803 Creates the directory specified by FILENAME, with permissions specified
1804 by MODE (as modified by umask). If it succeeds it returns 1, otherwise
1805 it returns 0 and sets C<$!> (errno).
1807 =item msgctl ID,CMD,ARG
1809 Calls the System V IPC function msgctl(2). If CMD is &IPC_STAT, then ARG
1810 must be a variable which will hold the returned msqid_ds structure.
1811 Returns like ioctl: the undefined value for error, "0 but true" for
1812 zero, or the actual return value otherwise.
1814 =item msgget KEY,FLAGS
1816 Calls the System V IPC function msgget(2). Returns the message queue id,
1817 or the undefined value if there is an error.
1819 =item msgsnd ID,MSG,FLAGS
1821 Calls the System V IPC function msgsnd to send the message MSG to the
1822 message queue ID. MSG must begin with the long integer message type,
1823 which may be created with C<pack("l", $type)>. Returns TRUE if
1824 successful, or FALSE if there is an error.
1826 =item msgrcv ID,VAR,SIZE,TYPE,FLAGS
1828 Calls the System V IPC function msgrcv to receive a message from
1829 message queue ID into variable VAR with a maximum message size of
1830 SIZE. Note that if a message is received, the message type will be the
1831 first thing in VAR, and the maximum length of VAR is SIZE plus the size
1832 of the message type. Returns TRUE if successful, or FALSE if there is
1837 A "my" declares the listed variables to be local (lexically) to the
1838 enclosing block, subroutine, C<eval>, or C<do/require/use>'d file. If
1839 more than one value is listed, the list must be placed in parentheses. See
1840 L<perlsub/"Private Variables via my()"> for details.
1846 The C<next> command is like the C<continue> statement in C; it starts
1847 the next iteration of the loop:
1849 LINE: while (<STDIN>) {
1850 next LINE if /^#/; # discard comments
1854 Note that if there were a C<continue> block on the above, it would get
1855 executed even on discarded lines. If the LABEL is omitted, the command
1856 refers to the innermost enclosing loop.
1858 =item no Module LIST
1860 See the "use" function, which "no" is the opposite of.
1866 Interprets EXPR as an octal string and returns the corresponding
1867 value. (If EXPR happens to start off with 0x, interprets it as
1868 a hex string instead.) The following will handle decimal, octal, and
1869 hex in the standard Perl or C notation:
1871 $val = oct($val) if $val =~ /^0/;
1873 If EXPR is omitted, uses $_. This function is commonly used when
1874 a string such as "644" needs to be converted into a file mode, for
1875 example. (Although perl will automatically convert strings into
1876 numbers as needed, this automatic conversion assumes base 10.)
1878 =item open FILEHANDLE,EXPR
1880 =item open FILEHANDLE
1882 Opens the file whose filename is given by EXPR, and associates it with
1883 FILEHANDLE. If FILEHANDLE is an expression, its value is used as the
1884 name of the real filehandle wanted. If EXPR is omitted, the scalar
1885 variable of the same name as the FILEHANDLE contains the filename.
1886 (Note that lexical variables--those declared with C<my>--will not work
1887 for this purpose; so if you're using C<my>, specify EXPR in your call
1890 If the filename begins with '<' or nothing, the file is opened for input.
1891 If the filename begins with '>', the file is truncated and opened for
1892 output. If the filename begins with '>>', the file is opened for
1893 appending. You can put a '+' in front of the '>' or '<' to indicate that
1894 you want both read and write access to the file; thus '+<' is almost
1895 always preferred for read/write updates--the '+>' mode would clobber the
1896 file first. The prefix and the filename may be separated with spaces.
1897 These various prefixes correspond to the fopen(3) modes of 'r', 'r+', 'w',
1898 'w+', 'a', and 'a+'.
1900 If the filename begins with "|", the filename is interpreted as a command
1901 to which output is to be piped, and if the filename ends with a "|", the
1902 filename is interpreted See L<perlipc/"Using open() for IPC"> for more
1903 examples of this. as command which pipes input to us. (You may not have
1904 a raw open() to a command that pipes both in I<and> out, but see
1905 L<IPC::Open2>, L<IPC::Open3>, and L<perlipc/"Bidirectional Communication">
1908 Opening '-' opens STDIN and opening 'E<gt>-' opens STDOUT. Open returns
1909 non-zero upon success, the undefined value otherwise. If the open
1910 involved a pipe, the return value happens to be the pid of the
1913 If you're unfortunate enough to be running Perl on a system that
1914 distinguishes between text files and binary files (modern operating
1915 systems don't care), then you should check out L</binmode> for tips for
1916 dealing with this. The key distinction between systems that need binmode
1917 and those that don't is their text file formats. Systems like Unix and
1918 Plan9 that delimit lines with a single character, and that encode that
1919 character in C as '\n', do not need C<binmode>. The rest need it.
1924 open ARTICLE or die "Can't find article $ARTICLE: $!\n";
1925 while (<ARTICLE>) {...
1927 open(LOG, '>>/usr/spool/news/twitlog'); # (log is reserved)
1929 open(DBASE, '+<dbase.mine'); # open for update
1931 open(ARTICLE, "caesar <$article |"); # decrypt article
1933 open(EXTRACT, "|sort >/tmp/Tmp$$"); # $$ is our process id
1935 # process argument list of files along with any includes
1937 foreach $file (@ARGV) {
1938 process($file, 'fh00');
1942 local($filename, $input) = @_;
1943 $input++; # this is a string increment
1944 unless (open($input, $filename)) {
1945 print STDERR "Can't open $filename: $!\n";
1949 while (<$input>) { # note use of indirection
1950 if (/^#include "(.*)"/) {
1951 process($1, $input);
1958 You may also, in the Bourne shell tradition, specify an EXPR beginning
1959 with "E<gt>&", in which case the rest of the string is interpreted as the
1960 name of a filehandle (or file descriptor, if numeric) which is to be
1961 duped and opened. You may use & after E<gt>, E<gt>E<gt>, E<lt>, +E<gt>,
1962 +E<gt>E<gt>, and +E<lt>. The
1963 mode you specify should match the mode of the original filehandle.
1964 (Duping a filehandle does not take into account any existing contents of
1966 Here is a script that saves, redirects, and restores STDOUT and
1970 open(SAVEOUT, ">&STDOUT");
1971 open(SAVEERR, ">&STDERR");
1973 open(STDOUT, ">foo.out") || die "Can't redirect stdout";
1974 open(STDERR, ">&STDOUT") || die "Can't dup stdout";
1976 select(STDERR); $| = 1; # make unbuffered
1977 select(STDOUT); $| = 1; # make unbuffered
1979 print STDOUT "stdout 1\n"; # this works for
1980 print STDERR "stderr 1\n"; # subprocesses too
1985 open(STDOUT, ">&SAVEOUT");
1986 open(STDERR, ">&SAVEERR");
1988 print STDOUT "stdout 2\n";
1989 print STDERR "stderr 2\n";
1992 If you specify "E<lt>&=N", where N is a number, then Perl will do an
1993 equivalent of C's fdopen() of that file descriptor; this is more
1994 parsimonious of file descriptors. For example:
1996 open(FILEHANDLE, "<&=$fd")
1998 If you open a pipe on the command "-", i.e., either "|-" or "-|", then
1999 there is an implicit fork done, and the return value of open is the pid
2000 of the child within the parent process, and 0 within the child
2001 process. (Use C<defined($pid)> to determine whether the open was successful.)
2002 The filehandle behaves normally for the parent, but i/o to that
2003 filehandle is piped from/to the STDOUT/STDIN of the child process.
2004 In the child process the filehandle isn't opened--i/o happens from/to
2005 the new STDOUT or STDIN. Typically this is used like the normal
2006 piped open when you want to exercise more control over just how the
2007 pipe command gets executed, such as when you are running setuid, and
2008 don't want to have to scan shell commands for metacharacters.
2009 The following pairs are more or less equivalent:
2011 open(FOO, "|tr '[a-z]' '[A-Z]'");
2012 open(FOO, "|-") || exec 'tr', '[a-z]', '[A-Z]';
2014 open(FOO, "cat -n '$file'|");
2015 open(FOO, "-|") || exec 'cat', '-n', $file;
2017 See L<perlipc/"Safe Pipe Opens"> for more examples of this.
2019 Explicitly closing any piped filehandle causes the parent process to
2020 wait for the child to finish, and returns the status value in C<$?>.
2021 Note: on any operation which may do a fork, unflushed buffers remain
2022 unflushed in both processes, which means you may need to set C<$|> to
2023 avoid duplicate output.
2025 Using the constructor from the IO::Handle package (or one of its
2026 subclasses, such as IO::File or IO::Socket),
2027 you can generate anonymous filehandles which have the scope of whatever
2028 variables hold references to them, and automatically close whenever
2029 and however you leave that scope:
2033 sub read_myfile_munged {
2035 my $handle = new IO::File;
2036 open($handle, "myfile") or die "myfile: $!";
2038 or return (); # Automatically closed here.
2039 mung $first or die "mung failed"; # Or here.
2040 return $first, <$handle> if $ALL; # Or here.
2044 The filename that is passed to open will have leading and trailing
2045 whitespace deleted. To open a file with arbitrary weird
2046 characters in it, it's necessary to protect any leading and trailing
2049 $file =~ s#^(\s)#./$1#;
2050 open(FOO, "< $file\0");
2052 If you want a "real" C open() (see L<open(2)> on your system), then
2053 you should use the sysopen() function. This is another way to
2054 protect your filenames from interpretation. For example:
2057 sysopen(HANDLE, $path, O_RDWR|O_CREAT|O_EXCL, 0700)
2058 or die "sysopen $path: $!";
2059 HANDLE->autoflush(1);
2060 HANDLE->print("stuff $$\n");
2062 print "File contains: ", <HANDLE>;
2064 See L</seek()> for some details about mixing reading and writing.
2066 =item opendir DIRHANDLE,EXPR
2068 Opens a directory named EXPR for processing by readdir(), telldir(),
2069 seekdir(), rewinddir(), and closedir(). Returns TRUE if successful.
2070 DIRHANDLEs have their own namespace separate from FILEHANDLEs.
2076 Returns the numeric ascii value of the first character of EXPR. If
2077 EXPR is omitted, uses $_. For the reverse, see L<chr>.
2079 =item pack TEMPLATE,LIST
2081 Takes an array or list of values and packs it into a binary structure,
2082 returning the string containing the structure. The TEMPLATE is a
2083 sequence of characters that give the order and type of values, as
2086 A An ascii string, will be space padded.
2087 a An ascii string, will be null padded.
2088 b A bit string (ascending bit order, like vec()).
2089 B A bit string (descending bit order).
2090 h A hex string (low nybble first).
2091 H A hex string (high nybble first).
2093 c A signed char value.
2094 C An unsigned char value.
2095 s A signed short value.
2096 S An unsigned short value.
2097 i A signed integer value.
2098 I An unsigned integer value.
2099 l A signed long value.
2100 L An unsigned long value.
2102 n A short in "network" order.
2103 N A long in "network" order.
2104 v A short in "VAX" (little-endian) order.
2105 V A long in "VAX" (little-endian) order.
2107 f A single-precision float in the native format.
2108 d A double-precision float in the native format.
2110 p A pointer to a null-terminated string.
2111 P A pointer to a structure (fixed-length string).
2113 u A uuencoded string.
2115 w A BER compressed integer. Bytes give an unsigned integer base
2116 128, most significant digit first, with as few digits as
2117 possible, and with the bit 8 of each byte except the last set
2122 @ Null fill to absolute position.
2124 Each letter may optionally be followed by a number which gives a repeat
2125 count. With all types except "a", "A", "b", "B", "h", "H", and "P" the
2126 pack function will gobble up that many values from the LIST. A * for the
2127 repeat count means to use however many items are left. The "a" and "A"
2128 types gobble just one value, but pack it as a string of length count,
2129 padding with nulls or spaces as necessary. (When unpacking, "A" strips
2130 trailing spaces and nulls, but "a" does not.) Likewise, the "b" and "B"
2131 fields pack a string that many bits long. The "h" and "H" fields pack a
2132 string that many nybbles long. The "P" packs a pointer to a structure of
2133 the size indicated by the length. Real numbers (floats and doubles) are
2134 in the native machine format only; due to the multiplicity of floating
2135 formats around, and the lack of a standard "network" representation, no
2136 facility for interchange has been made. This means that packed floating
2137 point data written on one machine may not be readable on another - even if
2138 both use IEEE floating point arithmetic (as the endian-ness of the memory
2139 representation is not part of the IEEE spec). Note that Perl uses doubles
2140 internally for all numeric calculation, and converting from double into
2141 float and thence back to double again will lose precision (i.e.,
2142 C<unpack("f", pack("f", $foo)>) will not in general equal $foo).
2146 $foo = pack("cccc",65,66,67,68);
2148 $foo = pack("c4",65,66,67,68);
2151 $foo = pack("ccxxcc",65,66,67,68);
2154 $foo = pack("s2",1,2);
2155 # "\1\0\2\0" on little-endian
2156 # "\0\1\0\2" on big-endian
2158 $foo = pack("a4","abcd","x","y","z");
2161 $foo = pack("aaaa","abcd","x","y","z");
2164 $foo = pack("a14","abcdefg");
2165 # "abcdefg\0\0\0\0\0\0\0"
2167 $foo = pack("i9pl", gmtime);
2168 # a real struct tm (on my system anyway)
2171 unpack("N", pack("B32", substr("0" x 32 . shift, -32)));
2174 The same template may generally also be used in the unpack function.
2176 =item package NAMESPACE
2178 Declares the compilation unit as being in the given namespace. The scope
2179 of the package declaration is from the declaration itself through the end of
2180 the enclosing block (the same scope as the local() operator). All further
2181 unqualified dynamic identifiers will be in this namespace. A package
2182 statement affects only dynamic variables--including those you've used
2183 local() on--but I<not> lexical variables created with my(). Typically it
2184 would be the first declaration in a file to be included by the C<require>
2185 or C<use> operator. You can switch into a package in more than one place;
2186 it influences merely which symbol table is used by the compiler for the
2187 rest of that block. You can refer to variables and filehandles in other
2188 packages by prefixing the identifier with the package name and a double
2189 colon: C<$Package::Variable>. If the package name is null, the C<main>
2190 package as assumed. That is, C<$::sail> is equivalent to C<$main::sail>.
2192 See L<perlmod/"Packages"> for more information about packages, modules,
2193 and classes. See L<perlsub> for other scoping issues.
2195 =item pipe READHANDLE,WRITEHANDLE
2197 Opens a pair of connected pipes like the corresponding system call.
2198 Note that if you set up a loop of piped processes, deadlock can occur
2199 unless you are very careful. In addition, note that Perl's pipes use
2200 stdio buffering, so you may need to set C<$|> to flush your WRITEHANDLE
2201 after each command, depending on the application.
2203 See L<IPC::Open2>, L<IPC::Open3>, and L<perlipc/"Bidirectional Communication">
2204 for examples of such things.
2210 Pops and returns the last value of the array, shortening the array by
2211 1. Has a similar effect to
2213 $tmp = $ARRAY[$#ARRAY--];
2215 If there are no elements in the array, returns the undefined value.
2216 If ARRAY is omitted, pops the
2217 @ARGV array in the main program, and the @_ array in subroutines, just
2224 Returns the offset of where the last C<m//g> search left off for the variable
2225 is in question ($_ is used when the variable is not specified). May be
2226 modified to change that offset. Such modification will also influence
2227 the C<\G> zero-width assertion in regular expressions. See L<perlre> and
2230 =item print FILEHANDLE LIST
2236 Prints a string or a comma-separated list of strings. Returns TRUE
2237 if successful. FILEHANDLE may be a scalar variable name, in which case
2238 the variable contains the name of or a reference to the filehandle, thus introducing one
2239 level of indirection. (NOTE: If FILEHANDLE is a variable and the next
2240 token is a term, it may be misinterpreted as an operator unless you
2241 interpose a + or put parentheses around the arguments.) If FILEHANDLE is
2242 omitted, prints by default to standard output (or to the last selected
2243 output channel--see L</select>). If LIST is also omitted, prints $_ to
2244 STDOUT. To set the default output channel to something other than
2245 STDOUT use the select operation. Note that, because print takes a
2246 LIST, anything in the LIST is evaluated in a list context, and any
2247 subroutine that you call will have one or more of its expressions
2248 evaluated in a list context. Also be careful not to follow the print
2249 keyword with a left parenthesis unless you want the corresponding right
2250 parenthesis to terminate the arguments to the print--interpose a + or
2251 put parentheses around all the arguments.
2253 Note that if you're storing FILEHANDLES in an array or other expression,
2254 you will have to use a block returning its value instead:
2256 print { $files[$i] } "stuff\n";
2257 print { $OK ? STDOUT : STDERR } "stuff\n";
2259 =item printf FILEHANDLE FORMAT, LIST
2261 =item printf FORMAT, LIST
2263 Equivalent to C<print FILEHANDLE sprintf(FORMAT, LIST)>. The first argument
2264 of the list will be interpreted as the printf format. If C<use locale> is
2265 in effect, the character used for the decimal point in formatted real numbers
2266 is affected by the LC_NUMERIC locale. See L<perllocale>.
2268 Don't fall into the trap of using a printf() when a simple
2269 print() would do. The print() is more efficient, and less
2272 =item prototype FUNCTION
2274 Returns the prototype of a function as a string (or C<undef> if the
2275 function has no prototype). FUNCTION is a reference to, or the name of,
2276 the function whose prototype you want to retrieve.
2278 =item push ARRAY,LIST
2280 Treats ARRAY as a stack, and pushes the values of LIST
2281 onto the end of ARRAY. The length of ARRAY increases by the length of
2282 LIST. Has the same effect as
2285 $ARRAY[++$#ARRAY] = $value;
2288 but is more efficient. Returns the new number of elements in the array.
2298 Generalized quotes. See L<perlop>.
2300 =item quotemeta EXPR
2304 Returns the value of EXPR with with all non-alphanumeric
2305 characters backslashed. (That is, all characters not matching
2306 C</[A-Za-z_0-9]/> will be preceded by a backslash in the
2307 returned string, regardless of any locale settings.)
2308 This is the internal function implementing
2309 the \Q escape in double-quoted strings.
2311 If EXPR is omitted, uses $_.
2317 Returns a random fractional number between 0 and the value of EXPR.
2318 (EXPR should be positive.) If EXPR is omitted, returns a value between
2319 0 and 1. Automatically calls srand() unless srand() has already been
2320 called. See also srand().
2322 (Note: If your rand function consistently returns numbers that are too
2323 large or too small, then your version of Perl was probably compiled
2324 with the wrong number of RANDBITS.)
2326 =item read FILEHANDLE,SCALAR,LENGTH,OFFSET
2328 =item read FILEHANDLE,SCALAR,LENGTH
2330 Attempts to read LENGTH bytes of data into variable SCALAR from the
2331 specified FILEHANDLE. Returns the number of bytes actually read, or
2332 undef if there was an error. SCALAR will be grown or shrunk to the
2333 length actually read. An OFFSET may be specified to place the read
2334 data at some other place than the beginning of the string. This call
2335 is actually implemented in terms of stdio's fread call. To get a true
2336 read system call, see sysread().
2338 =item readdir DIRHANDLE
2340 Returns the next directory entry for a directory opened by opendir().
2341 If used in a list context, returns all the rest of the entries in the
2342 directory. If there are no more entries, returns an undefined value in
2343 a scalar context or a null list in a list context.
2345 If you're planning to filetest the return values out of a readdir(), you'd
2346 better prepend the directory in question. Otherwise, because we didn't
2347 chdir() there, it would have been testing the wrong file.
2349 opendir(DIR, $some_dir) || die "can't opendir $some_dir: $!";
2350 @dots = grep { /^\./ && -f "$some_dir/$_" } readdir(DIR);
2357 Returns the value of a symbolic link, if symbolic links are
2358 implemented. If not, gives a fatal error. If there is some system
2359 error, returns the undefined value and sets C<$!> (errno). If EXPR is
2362 =item recv SOCKET,SCALAR,LEN,FLAGS
2364 Receives a message on a socket. Attempts to receive LENGTH bytes of
2365 data into variable SCALAR from the specified SOCKET filehandle.
2366 Actually does a C recvfrom(), so that it can returns the address of the
2367 sender. Returns the undefined value if there's an error. SCALAR will
2368 be grown or shrunk to the length actually read. Takes the same flags
2369 as the system call of the same name.
2370 See L<perlipc/"UDP: Message Passing"> for examples.
2376 The C<redo> command restarts the loop block without evaluating the
2377 conditional again. The C<continue> block, if any, is not executed. If
2378 the LABEL is omitted, the command refers to the innermost enclosing
2379 loop. This command is normally used by programs that want to lie to
2380 themselves about what was just input:
2382 # a simpleminded Pascal comment stripper
2383 # (warning: assumes no { or } in strings)
2384 LINE: while (<STDIN>) {
2385 while (s|({.*}.*){.*}|$1 |) {}
2390 if (/}/) { # end of comment?
2403 Returns a TRUE value if EXPR is a reference, FALSE otherwise. If EXPR
2404 is not specified, $_ will be used. The value returned depends on the
2405 type of thing the reference is a reference to.
2406 Builtin types include:
2415 If the referenced object has been blessed into a package, then that package
2416 name is returned instead. You can think of ref() as a typeof() operator.
2418 if (ref($r) eq "HASH") {
2419 print "r is a reference to a hash.\n";
2422 print "r is not a reference at all.\n";
2425 See also L<perlref>.
2427 =item rename OLDNAME,NEWNAME
2429 Changes the name of a file. Returns 1 for success, 0 otherwise. Will
2430 not work across file system boundaries.
2436 Demands some semantics specified by EXPR, or by $_ if EXPR is not
2437 supplied. If EXPR is numeric, demands that the current version of Perl
2438 (C<$]> or $PERL_VERSION) be equal or greater than EXPR.
2440 Otherwise, demands that a library file be included if it hasn't already
2441 been included. The file is included via the do-FILE mechanism, which is
2442 essentially just a variety of eval(). Has semantics similar to the following
2446 local($filename) = @_;
2447 return 1 if $INC{$filename};
2448 local($realfilename,$result);
2450 foreach $prefix (@INC) {
2451 $realfilename = "$prefix/$filename";
2452 if (-f $realfilename) {
2453 $result = do $realfilename;
2457 die "Can't find $filename in \@INC";
2460 die "$filename did not return true value" unless $result;
2461 $INC{$filename} = $realfilename;
2465 Note that the file will not be included twice under the same specified
2466 name. The file must return TRUE as the last statement to indicate
2467 successful execution of any initialization code, so it's customary to
2468 end such a file with "1;" unless you're sure it'll return TRUE
2469 otherwise. But it's better just to put the "C<1;>", in case you add more
2472 If EXPR is a bare word, the require assumes a "F<.pm>" extension and
2473 replaces "F<::>" with "F</>" in the filename for you,
2474 to make it easy to load standard modules. This form of loading of
2475 modules does not risk altering your namespace.
2477 For a yet-more-powerful import facility, see L</use> and
2484 Generally used in a C<continue> block at the end of a loop to clear
2485 variables and reset ?? searches so that they work again. The
2486 expression is interpreted as a list of single characters (hyphens
2487 allowed for ranges). All variables and arrays beginning with one of
2488 those letters are reset to their pristine state. If the expression is
2489 omitted, one-match searches (?pattern?) are reset to match again. Resets
2490 only variables or searches in the current package. Always returns
2493 reset 'X'; # reset all X variables
2494 reset 'a-z'; # reset lower case variables
2495 reset; # just reset ?? searches
2497 Resetting "A-Z" is not recommended because you'll wipe out your
2498 ARGV and ENV arrays. Resets only package variables--lexical variables
2499 are unaffected, but they clean themselves up on scope exit anyway,
2500 so you'll probably want to use them instead. See L</my>.
2504 Returns from a subroutine, eval or do with the value specified. (Note that
2505 in the absence of a return a subroutine or eval() will automatically
2506 return the value of the last expression evaluated.)
2510 In a list context, returns a list value consisting of the elements
2511 of LIST in the opposite order. In a scalar context, concatenates the
2512 elements of LIST, and returns a string value consisting of those bytes,
2513 but in the opposite order.
2515 print reverse <>; # line tac, last line first
2517 undef $/; # for efficiency of <>
2518 print scalar reverse <>; # byte tac, last line tsrif
2520 This operator is also handy for inverting a hash, although there are some
2521 caveats. If a value is duplicated in the original hash, only one of those
2522 can be represented as a key in the inverted hash. Also, this has to
2523 unwind one hash and build a whole new one, which may take some time
2526 %by_name = reverse %by_address; # Invert the hash
2528 =item rewinddir DIRHANDLE
2530 Sets the current position to the beginning of the directory for the
2531 readdir() routine on DIRHANDLE.
2533 =item rindex STR,SUBSTR,POSITION
2535 =item rindex STR,SUBSTR
2537 Works just like index except that it returns the position of the LAST
2538 occurrence of SUBSTR in STR. If POSITION is specified, returns the
2539 last occurrence at or before that position.
2541 =item rmdir FILENAME
2545 Deletes the directory specified by FILENAME if it is empty. If it
2546 succeeds it returns 1, otherwise it returns 0 and sets C<$!> (errno). If
2547 FILENAME is omitted, uses $_.
2551 The substitution operator. See L<perlop>.
2555 Forces EXPR to be interpreted in a scalar context and returns the value
2558 @counts = ( scalar @a, scalar @b, scalar @c );
2560 There is no equivalent operator to force an expression to
2561 be interpolated in a list context because it's in practice never
2562 needed. If you really wanted to do so, however, you could use
2563 the construction C<@{[ (some expression) ]}>, but usually a simple
2564 C<(some expression)> suffices.
2566 =item seek FILEHANDLE,POSITION,WHENCE
2568 Randomly positions the file pointer for FILEHANDLE, just like the fseek()
2569 call of stdio. FILEHANDLE may be an expression whose value gives the name
2570 of the filehandle. The values for WHENCE are 0 to set the file pointer to
2571 POSITION, 1 to set the it to current plus POSITION, and 2 to set it to EOF
2572 plus offset. You may use the values SEEK_SET, SEEK_CUR, and SEEK_END for
2573 this from POSIX module. Returns 1 upon success, 0 otherwise.
2575 On some systems you have to do a seek whenever you switch between reading
2576 and writing. Amongst other things, this may have the effect of calling
2577 stdio's clearerr(3). A "whence" of 1 (SEEK_CUR) is useful for not moving
2582 This is also useful for applications emulating C<tail -f>. Once you hit
2583 EOF on your read, and then sleep for a while, you might have to stick in a
2584 seek() to reset things. First the simple trick listed above to clear the
2585 filepointer. The seek() doesn't change the current position, but it
2586 I<does> clear the end-of-file condition on the handle, so that the next
2587 C<E<lt>FILEE<gt>> makes Perl try again to read something. We hope.
2589 If that doesn't work (some stdios are particularly cantankerous), then
2590 you may need something more like this:
2593 for ($curpos = tell(FILE); $_ = <FILE>; $curpos = tell(FILE)) {
2594 # search for some stuff and put it into files
2596 sleep($for_a_while);
2597 seek(FILE, $curpos, 0);
2600 =item seekdir DIRHANDLE,POS
2602 Sets the current position for the readdir() routine on DIRHANDLE. POS
2603 must be a value returned by telldir(). Has the same caveats about
2604 possible directory compaction as the corresponding system library
2607 =item select FILEHANDLE
2611 Returns the currently selected filehandle. Sets the current default
2612 filehandle for output, if FILEHANDLE is supplied. This has two
2613 effects: first, a C<write> or a C<print> without a filehandle will
2614 default to this FILEHANDLE. Second, references to variables related to
2615 output will refer to this output channel. For example, if you have to
2616 set the top of form format for more than one output channel, you might
2624 FILEHANDLE may be an expression whose value gives the name of the
2625 actual filehandle. Thus:
2627 $oldfh = select(STDERR); $| = 1; select($oldfh);
2629 Some programmers may prefer to think of filehandles as objects with
2630 methods, preferring to write the last example as:
2633 STDERR->autoflush(1);
2635 =item select RBITS,WBITS,EBITS,TIMEOUT
2637 This calls the select(2) system call with the bit masks specified, which
2638 can be constructed using fileno() and vec(), along these lines:
2640 $rin = $win = $ein = '';
2641 vec($rin,fileno(STDIN),1) = 1;
2642 vec($win,fileno(STDOUT),1) = 1;
2645 If you want to select on many filehandles you might wish to write a
2649 local(@fhlist) = split(' ',$_[0]);
2652 vec($bits,fileno($_),1) = 1;
2656 $rin = fhbits('STDIN TTY SOCK');
2660 ($nfound,$timeleft) =
2661 select($rout=$rin, $wout=$win, $eout=$ein, $timeout);
2663 or to block until something becomes ready just do this
2665 $nfound = select($rout=$rin, $wout=$win, $eout=$ein, undef);
2667 Most systems do not bother to return anything useful in $timeleft, so
2668 calling select() in a scalar context just returns $nfound.
2670 Any of the bit masks can also be undef. The timeout, if specified, is
2671 in seconds, which may be fractional. Note: not all implementations are
2672 capable of returning the $timeleft. If not, they always return
2673 $timeleft equal to the supplied $timeout.
2675 You can effect a sleep of 250 milliseconds this way:
2677 select(undef, undef, undef, 0.25);
2679 B<WARNING>: Do not attempt to mix buffered I/O (like read() or E<lt>FHE<gt>)
2680 with select(). You have to use sysread() instead.
2682 =item semctl ID,SEMNUM,CMD,ARG
2684 Calls the System V IPC function semctl. If CMD is &IPC_STAT or
2685 &GETALL, then ARG must be a variable which will hold the returned
2686 semid_ds structure or semaphore value array. Returns like ioctl: the
2687 undefined value for error, "0 but true" for zero, or the actual return
2690 =item semget KEY,NSEMS,FLAGS
2692 Calls the System V IPC function semget. Returns the semaphore id, or
2693 the undefined value if there is an error.
2695 =item semop KEY,OPSTRING
2697 Calls the System V IPC function semop to perform semaphore operations
2698 such as signaling and waiting. OPSTRING must be a packed array of
2699 semop structures. Each semop structure can be generated with
2700 C<pack("sss", $semnum, $semop, $semflag)>. The number of semaphore
2701 operations is implied by the length of OPSTRING. Returns TRUE if
2702 successful, or FALSE if there is an error. As an example, the
2703 following code waits on semaphore $semnum of semaphore id $semid:
2705 $semop = pack("sss", $semnum, -1, 0);
2706 die "Semaphore trouble: $!\n" unless semop($semid, $semop);
2708 To signal the semaphore, replace "-1" with "1".
2710 =item send SOCKET,MSG,FLAGS,TO
2712 =item send SOCKET,MSG,FLAGS
2714 Sends a message on a socket. Takes the same flags as the system call
2715 of the same name. On unconnected sockets you must specify a
2716 destination to send TO, in which case it does a C sendto(). Returns
2717 the number of characters sent, or the undefined value if there is an
2719 See L<perlipc/"UDP: Message Passing"> for examples.
2721 =item setpgrp PID,PGRP
2723 Sets the current process group for the specified PID, 0 for the current
2724 process. Will produce a fatal error if used on a machine that doesn't
2725 implement setpgrp(2). If the arguments are omitted, it defaults to
2726 0,0. Note that the POSIX version of setpgrp() does not accept any
2727 arguments, so only setpgrp 0,0 is portable.
2729 =item setpriority WHICH,WHO,PRIORITY
2731 Sets the current priority for a process, a process group, or a user.
2732 (See setpriority(2).) Will produce a fatal error if used on a machine
2733 that doesn't implement setpriority(2).
2735 =item setsockopt SOCKET,LEVEL,OPTNAME,OPTVAL
2737 Sets the socket option requested. Returns undefined if there is an
2738 error. OPTVAL may be specified as undef if you don't want to pass an
2745 Shifts the first value of the array off and returns it, shortening the
2746 array by 1 and moving everything down. If there are no elements in the
2747 array, returns the undefined value. If ARRAY is omitted, shifts the
2748 @ARGV array in the main program, and the @_ array in subroutines.
2749 (This is determined lexically.) See also unshift(), push(), and pop().
2750 Shift() and unshift() do the same thing to the left end of an array
2751 that pop() and push() do to the right end.
2753 =item shmctl ID,CMD,ARG
2755 Calls the System V IPC function shmctl. If CMD is &IPC_STAT, then ARG
2756 must be a variable which will hold the returned shmid_ds structure.
2757 Returns like ioctl: the undefined value for error, "0 but true" for
2758 zero, or the actual return value otherwise.
2760 =item shmget KEY,SIZE,FLAGS
2762 Calls the System V IPC function shmget. Returns the shared memory
2763 segment id, or the undefined value if there is an error.
2765 =item shmread ID,VAR,POS,SIZE
2767 =item shmwrite ID,STRING,POS,SIZE
2769 Reads or writes the System V shared memory segment ID starting at
2770 position POS for size SIZE by attaching to it, copying in/out, and
2771 detaching from it. When reading, VAR must be a variable which will
2772 hold the data read. When writing, if STRING is too long, only SIZE
2773 bytes are used; if STRING is too short, nulls are written to fill out
2774 SIZE bytes. Return TRUE if successful, or FALSE if there is an error.
2776 =item shutdown SOCKET,HOW
2778 Shuts down a socket connection in the manner indicated by HOW, which
2779 has the same interpretation as in the system call of the same name.
2785 Returns the sine of EXPR (expressed in radians). If EXPR is omitted,
2788 For the inverse sine operation, you may use the POSIX::sin()
2789 function, or use this relation:
2791 sub asin { atan2($_[0], sqrt(1 - $_[0] * $_[0])) }
2797 Causes the script to sleep for EXPR seconds, or forever if no EXPR.
2798 May be interrupted by sending the process a SIGALRM. Returns the
2799 number of seconds actually slept. You probably cannot mix alarm() and
2800 sleep() calls, because sleep() is often implemented using alarm().
2802 On some older systems, it may sleep up to a full second less than what
2803 you requested, depending on how it counts seconds. Most modern systems
2804 always sleep the full amount.
2806 For delays of finer granularity than one second, you may use Perl's
2807 syscall() interface to access setitimer(2) if your system supports it,
2808 or else see L</select()> below.
2810 See also the POSIX module's sigpause() function.
2812 =item socket SOCKET,DOMAIN,TYPE,PROTOCOL
2814 Opens a socket of the specified kind and attaches it to filehandle
2815 SOCKET. DOMAIN, TYPE, and PROTOCOL are specified the same as for the
2816 system call of the same name. You should "use Socket;" first to get
2817 the proper definitions imported. See the example in L<perlipc/"Sockets: Client/Server Communication">.
2819 =item socketpair SOCKET1,SOCKET2,DOMAIN,TYPE,PROTOCOL
2821 Creates an unnamed pair of sockets in the specified domain, of the
2822 specified type. DOMAIN, TYPE, and PROTOCOL are specified the same as
2823 for the system call of the same name. If unimplemented, yields a fatal
2824 error. Returns TRUE if successful.
2826 =item sort SUBNAME LIST
2828 =item sort BLOCK LIST
2832 Sorts the LIST and returns the sorted list value. If SUBNAME or BLOCK
2833 is omitted, sorts in standard string comparison order. If SUBNAME is
2834 specified, it gives the name of a subroutine that returns an integer
2835 less than, equal to, or greater than 0, depending on how the elements
2836 of the array are to be ordered. (The C<E<lt>=E<gt>> and C<cmp>
2837 operators are extremely useful in such routines.) SUBNAME may be a
2838 scalar variable name, in which case the value provides the name of the
2839 subroutine to use. In place of a SUBNAME, you can provide a BLOCK as
2840 an anonymous, in-line sort subroutine.
2842 In the interests of efficiency the normal calling code for subroutines is
2843 bypassed, with the following effects: the subroutine may not be a
2844 recursive subroutine, and the two elements to be compared are passed into
2845 the subroutine not via @_ but as the package global variables $a and
2846 $b (see example below). They are passed by reference, so don't
2847 modify $a and $b. And don't try to declare them as lexicals either.
2849 You also cannot exit out of the sort block or subroutine using any of the
2850 loop control operators described in L<perlsyn> or with goto().
2852 When C<use locale> is in effect, C<sort LIST> sorts LIST according to the
2853 current collation locale. See L<perllocale>.
2858 @articles = sort @files;
2860 # same thing, but with explicit sort routine
2861 @articles = sort {$a cmp $b} @files;
2863 # now case-insensitively
2864 @articles = sort { uc($a) cmp uc($b)} @files;
2866 # same thing in reversed order
2867 @articles = sort {$b cmp $a} @files;
2869 # sort numerically ascending
2870 @articles = sort {$a <=> $b} @files;
2872 # sort numerically descending
2873 @articles = sort {$b <=> $a} @files;
2875 # sort using explicit subroutine name
2877 $age{$a} <=> $age{$b}; # presuming numeric
2879 @sortedclass = sort byage @class;
2881 # this sorts the %age hash by value instead of key
2882 # using an in-line function
2883 @eldest = sort { $age{$b} <=> $age{$a} } keys %age;
2885 sub backwards { $b cmp $a; }
2886 @harry = ('dog','cat','x','Cain','Abel');
2887 @george = ('gone','chased','yz','Punished','Axed');
2889 # prints AbelCaincatdogx
2890 print sort backwards @harry;
2891 # prints xdogcatCainAbel
2892 print sort @george, 'to', @harry;
2893 # prints AbelAxedCainPunishedcatchaseddoggonetoxyz
2895 # inefficiently sort by descending numeric compare using
2896 # the first integer after the first = sign, or the
2897 # whole record case-insensitively otherwise
2900 ($b =~ /=(\d+)/)[0] <=> ($a =~ /=(\d+)/)[0]
2905 # same thing, but much more efficiently;
2906 # we'll build auxiliary indices instead
2910 push @nums, /=(\d+)/;
2915 $nums[$b] <=> $nums[$a]
2917 $caps[$a] cmp $caps[$b]
2921 # same thing using a Schwartzian Transform (no temps)
2922 @new = map { $_->[0] }
2923 sort { $b->[1] <=> $a->[1]
2926 } map { [$_, /=(\d+)/, uc($_)] } @old;
2928 If you're using strict, you I<MUST NOT> declare $a
2929 and $b as lexicals. They are package globals. That means
2930 if you're in the C<main> package, it's
2932 @articles = sort {$main::b <=> $main::a} @files;
2936 @articles = sort {$::b <=> $::a} @files;
2938 but if you're in the C<FooPack> package, it's
2940 @articles = sort {$FooPack::b <=> $FooPack::a} @files;
2942 The comparison function is required to behave. If it returns
2943 inconsistent results (sometimes saying $x[1] is less than $x[2] and
2944 sometimes saying the opposite, for example) the Perl interpreter will
2945 probably crash and dump core. This is entirely due to and dependent
2946 upon your system's qsort(3) library routine; this routine often avoids
2947 sanity checks in the interest of speed.
2949 =item splice ARRAY,OFFSET,LENGTH,LIST
2951 =item splice ARRAY,OFFSET,LENGTH
2953 =item splice ARRAY,OFFSET
2955 Removes the elements designated by OFFSET and LENGTH from an array, and
2956 replaces them with the elements of LIST, if any. Returns the elements
2957 removed from the array. The array grows or shrinks as necessary. If
2958 LENGTH is omitted, removes everything from OFFSET onward. The
2959 following equivalences hold (assuming C<$[ == 0>):
2961 push(@a,$x,$y) splice(@a,$#a+1,0,$x,$y)
2962 pop(@a) splice(@a,-1)
2963 shift(@a) splice(@a,0,1)
2964 unshift(@a,$x,$y) splice(@a,0,0,$x,$y)
2965 $a[$x] = $y splice(@a,$x,1,$y);
2967 Example, assuming array lengths are passed before arrays:
2969 sub aeq { # compare two list values
2970 local(@a) = splice(@_,0,shift);
2971 local(@b) = splice(@_,0,shift);
2972 return 0 unless @a == @b; # same len?
2974 return 0 if pop(@a) ne pop(@b);
2978 if (&aeq($len,@foo[1..$len],0+@bar,@bar)) { ... }
2980 =item split /PATTERN/,EXPR,LIMIT
2982 =item split /PATTERN/,EXPR
2984 =item split /PATTERN/
2988 Splits a string into an array of strings, and returns it.
2990 If not in a list context, returns the number of fields found and splits into
2991 the @_ array. (In a list context, you can force the split into @_ by
2992 using C<??> as the pattern delimiters, but it still returns the array
2993 value.) The use of implicit split to @_ is deprecated, however.
2995 If EXPR is omitted, splits the $_ string. If PATTERN is also omitted,
2996 splits on whitespace (after skipping any leading whitespace). Anything
2997 matching PATTERN is taken to be a delimiter separating the fields. (Note
2998 that the delimiter may be longer than one character.) If LIMIT is
2999 specified and is not negative, splits into no more than that many fields
3000 (though it may split into fewer). If LIMIT is unspecified, trailing null
3001 fields are stripped (which potential users of pop() would do well to
3002 remember). If LIMIT is negative, it is treated as if an arbitrarily large
3003 LIMIT had been specified.
3005 A pattern matching the null string (not to be confused with
3006 a null pattern C<//>, which is just one member of the set of patterns
3007 matching a null string) will split the value of EXPR into separate
3008 characters at each point it matches that way. For example:
3010 print join(':', split(/ */, 'hi there'));
3012 produces the output 'h:i:t:h:e:r:e'.
3014 The LIMIT parameter can be used to split a line partially
3016 ($login, $passwd, $remainder) = split(/:/, $_, 3);
3018 When assigning to a list, if LIMIT is omitted, Perl supplies a LIMIT
3019 one larger than the number of variables in the list, to avoid
3020 unnecessary work. For the list above LIMIT would have been 4 by
3021 default. In time critical applications it behooves you not to split
3022 into more fields than you really need.
3024 If the PATTERN contains parentheses, additional array elements are
3025 created from each matching substring in the delimiter.
3027 split(/([,-])/, "1-10,20", 3);
3029 produces the list value
3031 (1, '-', 10, ',', 20)
3033 If you had the entire header of a normal Unix email message in $header,
3034 you could split it up into fields and their values this way:
3036 $header =~ s/\n\s+/ /g; # fix continuation lines
3037 %hdrs = (UNIX_FROM => split /^(.*?):\s*/m, $header);
3039 The pattern C</PATTERN/> may be replaced with an expression to specify
3040 patterns that vary at runtime. (To do runtime compilation only once,
3041 use C</$variable/o>.)
3043 As a special case, specifying a PATTERN of space (C<' '>) will split on
3044 white space just as split with no arguments does. Thus, split(' ') can
3045 be used to emulate B<awk>'s default behavior, whereas C<split(/ /)>
3046 will give you as many null initial fields as there are leading spaces.
3047 A split on /\s+/ is like a split(' ') except that any leading
3048 whitespace produces a null first field. A split with no arguments
3049 really does a C<split(' ', $_)> internally.
3053 open(passwd, '/etc/passwd');
3055 ($login, $passwd, $uid, $gid, $gcos,
3056 $home, $shell) = split(/:/);
3060 (Note that $shell above will still have a newline on it. See L</chop>,
3061 L</chomp>, and L</join>.)
3063 =item sprintf FORMAT, LIST
3065 Returns a string formatted by the usual printf conventions of the C
3066 language. See L<sprintf(3)> or L<printf(3)> on your system for details.
3067 (The * character for an indirectly specified length is not
3068 supported, but you can get the same effect by interpolating a variable
3069 into the pattern.) If C<use locale> is
3070 in effect, the character used for the decimal point in formatted real numbers
3071 is affected by the LC_NUMERIC locale. See L<perllocale>.
3072 Some C libraries' implementations of sprintf() can
3073 dump core when fed ludicrous arguments.
3079 Return the square root of EXPR. If EXPR is omitted, returns square
3086 Sets the random number seed for the C<rand> operator. If EXPR is
3087 omitted, uses a semi-random value based on the current time and process
3088 ID, among other things. In versions of Perl prior to 5.004 the default
3089 seed was just the current time(). This isn't a particularly good seed,
3090 so many old programs supply their own seed value (often C<time ^ $$> or
3091 C<time ^ ($$ + ($$ << 15))>), but that isn't necessary any more.
3093 In fact, it's usually not necessary to call srand() at all, because if
3094 it is not called explicitly, it is called implicitly at the first use of
3095 the C<rand> operator. However, this was not the case in version of Perl
3096 before 5.004, so if your script will run under older Perl versions, it
3097 should call srand().
3099 Note that you need something much more random than the default seed for
3100 cryptographic purposes. Checksumming the compressed output of one or more
3101 rapidly changing operating system status programs is the usual method. For
3104 srand (time ^ $$ ^ unpack "%L*", `ps axww | gzip`);
3106 If you're particularly concerned with this, see the Math::TrulyRandom
3109 Do I<not> call srand() multiple times in your program unless you know
3110 exactly what you're doing and why you're doing it. The point of the
3111 function is to "seed" the rand() function so that rand() can produce
3112 a different sequence each time you run your program. Just do it once at the
3113 top of your program, or you I<won't> get random numbers out of rand()!
3115 Frequently called programs (like CGI scripts) that simply use
3119 for a seed can fall prey to the mathematical property that
3123 one-third of the time. So don't do that.
3125 =item stat FILEHANDLE
3131 Returns a 13-element array giving the status info for a file, either the
3132 file opened via FILEHANDLE, or named by EXPR. If EXPR is omitted, it
3133 stats $_. Returns a null list if the stat fails. Typically used as
3137 ($dev,$ino,$mode,$nlink,$uid,$gid,$rdev,$size,
3138 $atime,$mtime,$ctime,$blksize,$blocks)
3141 Not all fields are supported on all filesystem types. Here are the
3142 meaning of the fields:
3144 dev device number of filesystem
3146 mode file mode (type and permissions)
3147 nlink number of (hard) links to the file
3148 uid numeric user ID of file's owner
3149 gid numeric group ID of file's owner
3150 rdev the device identifier (special files only)
3151 size total size of file, in bytes
3152 atime last access time since the epoch
3153 mtime last modify time since the epoch
3154 ctime inode change time (NOT creation time!) since the epoch
3155 blksize preferred block size for file system I/O
3156 blocks actual number of blocks allocated
3158 (The epoch was at 00:00 January 1, 1970 GMT.)
3160 If stat is passed the special filehandle consisting of an underline, no
3161 stat is done, but the current contents of the stat structure from the
3162 last stat or filetest are returned. Example:
3164 if (-x $file && (($d) = stat(_)) && $d < 0) {
3165 print "$file is executable NFS file\n";
3168 (This works on machines only for which the device number is negative under NFS.)
3174 Takes extra time to study SCALAR (C<$_> if unspecified) in anticipation of
3175 doing many pattern matches on the string before it is next modified.
3176 This may or may not save time, depending on the nature and number of
3177 patterns you are searching on, and on the distribution of character
3178 frequencies in the string to be searched--you probably want to compare
3179 run times with and without it to see which runs faster. Those loops
3180 which scan for many short constant strings (including the constant
3181 parts of more complex patterns) will benefit most. You may have only
3182 one study active at a time--if you study a different scalar the first
3183 is "unstudied". (The way study works is this: a linked list of every
3184 character in the string to be searched is made, so we know, for
3185 example, where all the 'k' characters are. From each search string,
3186 the rarest character is selected, based on some static frequency tables
3187 constructed from some C programs and English text. Only those places
3188 that contain this "rarest" character are examined.)
3190 For example, here is a loop which inserts index producing entries
3191 before any line containing a certain pattern:
3195 print ".IX foo\n" if /\bfoo\b/;
3196 print ".IX bar\n" if /\bbar\b/;
3197 print ".IX blurfl\n" if /\bblurfl\b/;
3202 In searching for /\bfoo\b/, only those locations in $_ that contain "f"
3203 will be looked at, because "f" is rarer than "o". In general, this is
3204 a big win except in pathological cases. The only question is whether
3205 it saves you more time than it took to build the linked list in the
3208 Note that if you have to look for strings that you don't know till
3209 runtime, you can build an entire loop as a string and eval that to
3210 avoid recompiling all your patterns all the time. Together with
3211 undefining $/ to input entire files as one record, this can be very
3212 fast, often faster than specialized programs like fgrep(1). The following
3213 scans a list of files (C<@files>) for a list of words (C<@words>), and prints
3214 out the names of those files that contain a match:
3216 $search = 'while (<>) { study;';
3217 foreach $word (@words) {
3218 $search .= "++\$seen{\$ARGV} if /\\b$word\\b/;\n";
3223 eval $search; # this screams
3224 $/ = "\n"; # put back to normal input delimiter
3225 foreach $file (sort keys(%seen)) {
3233 =item sub NAME BLOCK
3235 This is subroutine definition, not a real function I<per se>. With just a
3236 NAME (and possibly prototypes), it's just a forward declaration. Without
3237 a NAME, it's an anonymous function declaration, and does actually return a
3238 value: the CODE ref of the closure you just created. See L<perlsub> and
3239 L<perlref> for details.
3241 =item substr EXPR,OFFSET,LEN
3243 =item substr EXPR,OFFSET
3245 Extracts a substring out of EXPR and returns it. First character is at
3246 offset 0, or whatever you've set C<$[> to (but don't do that).
3247 If OFFSET is negative, starts
3248 that far from the end of the string. If LEN is omitted, returns
3249 everything to the end of the string. If LEN is negative, leaves that
3250 many characters off the end of the string.
3252 You can use the substr() function
3253 as an lvalue, in which case EXPR must be an lvalue. If you assign
3254 something shorter than LEN, the string will shrink, and if you assign
3255 something longer than LEN, the string will grow to accommodate it. To
3256 keep the string the same length you may need to pad or chop your value
3259 =item symlink OLDFILE,NEWFILE
3261 Creates a new filename symbolically linked to the old filename.
3262 Returns 1 for success, 0 otherwise. On systems that don't support
3263 symbolic links, produces a fatal error at run time. To check for that,
3266 $symlink_exists = (eval 'symlink("","");', $@ eq '');
3270 Calls the system call specified as the first element of the list,
3271 passing the remaining elements as arguments to the system call. If
3272 unimplemented, produces a fatal error. The arguments are interpreted
3273 as follows: if a given argument is numeric, the argument is passed as
3274 an int. If not, the pointer to the string value is passed. You are
3275 responsible to make sure a string is pre-extended long enough to
3276 receive any result that might be written into a string. If your
3277 integer arguments are not literals and have never been interpreted in a
3278 numeric context, you may need to add 0 to them to force them to look
3281 require 'syscall.ph'; # may need to run h2ph
3282 syscall(&SYS_write, fileno(STDOUT), "hi there\n", 9);
3284 Note that Perl supports passing of up to only 14 arguments to your system call,
3285 which in practice should usually suffice.
3287 =item sysopen FILEHANDLE,FILENAME,MODE
3289 =item sysopen FILEHANDLE,FILENAME,MODE,PERMS
3291 Opens the file whose filename is given by FILENAME, and associates it
3292 with FILEHANDLE. If FILEHANDLE is an expression, its value is used as
3293 the name of the real filehandle wanted. This function calls the
3294 underlying operating system's C<open> function with the parameters
3295 FILENAME, MODE, PERMS.
3297 The possible values and flag bits of the MODE parameter are
3298 system-dependent; they are available via the standard module C<Fcntl>.
3299 However, for historical reasons, some values are universal: zero means
3300 read-only, one means write-only, and two means read/write.
3302 If the file named by FILENAME does not exist and the C<open> call
3303 creates it (typically because MODE includes the O_CREAT flag), then
3304 the value of PERMS specifies the permissions of the newly created
3305 file. If PERMS is omitted, the default value is 0666, which allows
3306 read and write for all. This default is reasonable: see C<umask>.
3308 The IO::File module provides a more object-oriented approach, if you're
3309 into that kind of thing.
3311 =item sysread FILEHANDLE,SCALAR,LENGTH,OFFSET
3313 =item sysread FILEHANDLE,SCALAR,LENGTH
3315 Attempts to read LENGTH bytes of data into variable SCALAR from the
3316 specified FILEHANDLE, using the system call read(2). It bypasses
3317 stdio, so mixing this with other kinds of reads may cause confusion.
3318 Returns the number of bytes actually read, or undef if there was an
3319 error. SCALAR will be grown or shrunk so that the last byte actually
3320 read is the last byte of the scalar after the read.
3322 An OFFSET may be specified to place the read data at some place in the
3323 string other than the beginning. A negative OFFSET specifies
3324 placement at that many bytes counting backwards from the end of the
3325 string. A positive OFFSET greater than the length of SCALAR results
3326 in the string being padded to the required size with "\0" bytes before
3327 the result of the read is appended.
3331 Does exactly the same thing as "exec LIST" except that a fork is done
3332 first, and the parent process waits for the child process to complete.
3333 Note that argument processing varies depending on the number of
3334 arguments. The return value is the exit status of the program as
3335 returned by the wait() call. To get the actual exit value divide by
3336 256. See also L</exec>. This is I<NOT> what you want to use to capture
3337 the output from a command, for that you should use merely back-ticks or
3338 qx//, as described in L<perlop/"`STRING`">.
3340 Because system() and back-ticks block SIGINT and SIGQUIT, killing the
3341 program they're running doesn't actually interrupt your program.
3343 @args = ("command", "arg1", "arg2");
3345 or die "system @args failed: $?"
3347 Here's a more elaborate example of analysing the return value from
3348 system() on a UNIX system to check for all possibilities, including for
3349 signals and coredumps.
3351 $rc = 0xffff & system @args;
3352 printf "system(%s) returned %#04x: ", "@args", $rc;
3354 print "ran with normal exit\n";
3356 elsif ($rc == 0xff00) {
3357 print "command failed: $!\n";
3359 elsif ($rc > 0x80) {
3361 print "ran with non-zero exit status $rc\n";
3367 print "coredump from ";
3369 print "signal $rc\n"
3373 =item syswrite FILEHANDLE,SCALAR,LENGTH,OFFSET
3375 =item syswrite FILEHANDLE,SCALAR,LENGTH
3377 Attempts to write LENGTH bytes of data from variable SCALAR to the
3378 specified FILEHANDLE, using the system call write(2). It bypasses
3379 stdio, so mixing this with prints may cause confusion. Returns the
3380 number of bytes actually written, or undef if there was an error.
3381 If the length is greater than the available data, only as much data as
3382 is available will be written.
3384 An OFFSET may be specified to write the data from some part of the
3385 string other than the beginning. A negative OFFSET specifies writing
3386 from that many bytes counting backwards from the end of the string.
3388 =item tell FILEHANDLE
3392 Returns the current file position for FILEHANDLE. FILEHANDLE may be an
3393 expression whose value gives the name of the actual filehandle. If
3394 FILEHANDLE is omitted, assumes the file last read.
3396 =item telldir DIRHANDLE
3398 Returns the current position of the readdir() routines on DIRHANDLE.
3399 Value may be given to seekdir() to access a particular location in a
3400 directory. Has the same caveats about possible directory compaction as
3401 the corresponding system library routine.
3403 =item tie VARIABLE,CLASSNAME,LIST
3405 This function binds a variable to a package class that will provide the
3406 implementation for the variable. VARIABLE is the name of the variable
3407 to be enchanted. CLASSNAME is the name of a class implementing objects
3408 of correct type. Any additional arguments are passed to the "new"
3409 method of the class (meaning TIESCALAR, TIEARRAY, or TIEHASH).
3410 Typically these are arguments such as might be passed to the dbm_open()
3411 function of C. The object returned by the "new" method is also
3412 returned by the tie() function, which would be useful if you want to
3413 access other methods in CLASSNAME.
3415 Note that functions such as keys() and values() may return huge array
3416 values when used on large objects, like DBM files. You may prefer to
3417 use the each() function to iterate over such. Example:
3419 # print out history file offsets
3421 tie(%HIST, 'NDBM_File', '/usr/lib/news/history', 1, 0);
3422 while (($key,$val) = each %HIST) {
3423 print $key, ' = ', unpack('L',$val), "\n";
3427 A class implementing a hash should have the following methods:
3429 TIEHASH classname, LIST
3432 STORE this, key, value
3436 NEXTKEY this, lastkey
3438 A class implementing an ordinary array should have the following methods:
3440 TIEARRAY classname, LIST
3443 STORE this, key, value
3446 A class implementing a scalar should have the following methods:
3448 TIESCALAR classname, LIST
3453 Unlike dbmopen(), the tie() function will not use or require a module
3454 for you--you need to do that explicitly yourself. See L<DB_File>
3455 or the F<Config> module for interesting tie() implementations.
3459 Returns a reference to the object underlying VARIABLE (the same value
3460 that was originally returned by the tie() call which bound the variable
3461 to a package.) Returns the undefined value if VARIABLE isn't tied to a
3466 Returns the number of non-leap seconds since whatever time the system
3467 considers to be the epoch (that's 00:00:00, January 1, 1904 for MacOS,
3468 and 00:00:00 UTC, January 1, 1970 for most other systems).
3469 Suitable for feeding to gmtime() and localtime().
3473 Returns a four-element array giving the user and system times, in
3474 seconds, for this process and the children of this process.
3476 ($user,$system,$cuser,$csystem) = times;
3480 The translation operator. See L<perlop>.
3482 =item truncate FILEHANDLE,LENGTH
3484 =item truncate EXPR,LENGTH
3486 Truncates the file opened on FILEHANDLE, or named by EXPR, to the
3487 specified length. Produces a fatal error if truncate isn't implemented
3494 Returns an uppercased version of EXPR. This is the internal function
3495 implementing the \U escape in double-quoted strings.
3496 Respects current LC_CTYPE locale if C<use locale> in force. See L<perllocale>.
3498 If EXPR is omitted, uses $_.
3504 Returns the value of EXPR with the first character uppercased. This is
3505 the internal function implementing the \u escape in double-quoted strings.
3506 Respects current LC_CTYPE locale if C<use locale> in force. See L<perllocale>.
3508 If EXPR is omitted, uses $_.
3514 Sets the umask for the process to EXPR and returns the previous value.
3515 If EXPR is omitted, merely returns the current umask. Remember that a
3516 umask is a number, usually given in octal; it is I<not> a string of octal
3517 digits. See also L<oct>, if all you have is a string.
3523 Undefines the value of EXPR, which must be an lvalue. Use on only a
3524 scalar value, an entire array or hash, or a subroutine name (using
3525 "&"). (Using undef() will probably not do what you expect on most
3526 predefined variables or DBM list values, so don't do that.) Always
3527 returns the undefined value. You can omit the EXPR, in which case
3528 nothing is undefined, but you still get an undefined value that you
3529 could, for instance, return from a subroutine, assign to a variable or
3530 pass as a parameter. Examples:
3533 undef $bar{'blurfl'}; # Compare to: delete $bar{'blurfl'};
3537 return (wantarray ? () : undef) if $they_blew_it;
3538 select undef, undef, undef, 0.25;
3539 ($a, $b, undef, $c) = &foo; # Ignore third value returned
3545 Deletes a list of files. Returns the number of files successfully
3548 $cnt = unlink 'a', 'b', 'c';
3552 Note: unlink will not delete directories unless you are superuser and
3553 the B<-U> flag is supplied to Perl. Even if these conditions are
3554 met, be warned that unlinking a directory can inflict damage on your
3555 filesystem. Use rmdir instead.
3557 If LIST is omitted, uses $_.
3559 =item unpack TEMPLATE,EXPR
3561 Unpack does the reverse of pack: it takes a string representing a
3562 structure and expands it out into a list value, returning the array
3563 value. (In a scalar context, it returns merely the first value
3564 produced.) The TEMPLATE has the same format as in the pack function.
3565 Here's a subroutine that does substring:
3568 local($what,$where,$howmuch) = @_;
3569 unpack("x$where a$howmuch", $what);
3574 sub ordinal { unpack("c",$_[0]); } # same as ord()
3576 In addition, you may prefix a field with a %E<lt>numberE<gt> to indicate that
3577 you want a E<lt>numberE<gt>-bit checksum of the items instead of the items
3578 themselves. Default is a 16-bit checksum. For example, the following
3579 computes the same number as the System V sum program:
3582 $checksum += unpack("%16C*", $_);
3586 The following efficiently counts the number of set bits in a bit vector:
3588 $setbits = unpack("%32b*", $selectmask);
3590 =item untie VARIABLE
3592 Breaks the binding between a variable and a package. (See tie().)
3594 =item unshift ARRAY,LIST
3596 Does the opposite of a C<shift>. Or the opposite of a C<push>,
3597 depending on how you look at it. Prepends list to the front of the
3598 array, and returns the new number of elements in the array.
3600 unshift(ARGV, '-e') unless $ARGV[0] =~ /^-/;
3602 Note the LIST is prepended whole, not one element at a time, so the
3603 prepended elements stay in the same order. Use reverse to do the
3606 =item use Module LIST
3610 =item use Module VERSION LIST
3614 Imports some semantics into the current package from the named module,
3615 generally by aliasing certain subroutine or variable names into your
3616 package. It is exactly equivalent to
3618 BEGIN { require Module; import Module LIST; }
3620 except that Module I<must> be a bare word.
3622 If the first argument to C<use> is a number, it is treated as a version
3623 number instead of a module name. If the version of the Perl interpreter
3624 is less than VERSION, then an error message is printed and Perl exits
3625 immediately. This is often useful if you need to check the current
3626 Perl version before C<use>ing library modules which have changed in
3627 incompatible ways from older versions of Perl. (We try not to do
3628 this more than we have to.)
3630 The BEGIN forces the require and import to happen at compile time. The
3631 require makes sure the module is loaded into memory if it hasn't been
3632 yet. The import is not a builtin--it's just an ordinary static method
3633 call into the "Module" package to tell the module to import the list of
3634 features back into the current package. The module can implement its
3635 import method any way it likes, though most modules just choose to
3636 derive their import method via inheritance from the Exporter class that
3637 is defined in the Exporter module. See L<Exporter>. If no import
3638 method can be found then the error is currently silently ignored. This
3639 may change to a fatal error in a future version.
3641 If you don't want your namespace altered, explicitly supply an empty list:
3645 That is exactly equivalent to
3647 BEGIN { require Module; }
3649 If the VERSION argument is present between Module and LIST, then the
3650 C<use> will call the VERSION method in class Module with the given
3651 version as an argument. The default VERSION method, inherited from
3652 the Universal class, croaks if the given version is larger than the
3653 value of the variable $Module::VERSION. (Note that there is not a
3654 comma after VERSION!)
3656 Because this is a wide-open interface, pragmas (compiler directives)
3657 are also implemented this way. Currently implemented pragmas are:
3661 use sigtrap qw(SEGV BUS);
3662 use strict qw(subs vars refs);
3663 use subs qw(afunc blurfl);
3665 These pseudo-modules import semantics into the current block scope, unlike
3666 ordinary modules, which import symbols into the current package (which are
3667 effective through the end of the file).
3669 There's a corresponding "no" command that unimports meanings imported
3670 by use, i.e., it calls C<unimport Module LIST> instead of C<import>.
3675 If no unimport method can be found the call fails with a fatal error.
3677 See L<perlmod> for a list of standard modules and pragmas.
3681 Changes the access and modification times on each file of a list of
3682 files. The first two elements of the list must be the NUMERICAL access
3683 and modification times, in that order. Returns the number of files
3684 successfully changed. The inode modification time of each file is set
3685 to the current time. Example of a "touch" command:
3689 utime $now, $now, @ARGV;
3693 Returns a normal array consisting of all the values of the named hash.
3694 (In a scalar context, returns the number of values.) The values are
3695 returned in an apparently random order, but it is the same order as either
3696 the keys() or each() function would produce on the same hash. As a side
3697 effect, it resets HASH's iterator. See also keys(), each(), and sort().
3699 =item vec EXPR,OFFSET,BITS
3701 Treats the string in EXPR as a vector of unsigned integers, and
3702 returns the value of the bit field specified by OFFSET. BITS specifies
3703 the number of bits that are reserved for each entry in the bit
3704 vector. This must be a power of two from 1 to 32. vec() may also be
3705 assigned to, in which case parentheses are needed to give the expression
3706 the correct precedence as in
3708 vec($image, $max_x * $x + $y, 8) = 3;
3710 Vectors created with vec() can also be manipulated with the logical
3711 operators |, &, and ^, which will assume a bit vector operation is
3712 desired when both operands are strings.
3714 To transform a bit vector into a string or array of 0's and 1's, use these:
3716 $bits = unpack("b*", $vector);
3717 @bits = split(//, unpack("b*", $vector));
3719 If you know the exact length in bits, it can be used in place of the *.
3723 Waits for a child process to terminate and returns the pid of the
3724 deceased process, or -1 if there are no child processes. The status is
3727 =item waitpid PID,FLAGS
3729 Waits for a particular child process to terminate and returns the pid
3730 of the deceased process, or -1 if there is no such child process. The
3731 status is returned in C<$?>. If you say
3733 use POSIX ":sys_wait_h";
3735 waitpid(-1,&WNOHANG);
3737 then you can do a non-blocking wait for any process. Non-blocking wait
3738 is available on machines supporting either the waitpid(2) or
3739 wait4(2) system calls. However, waiting for a particular pid with
3740 FLAGS of 0 is implemented everywhere. (Perl emulates the system call
3741 by remembering the status values of processes that have exited but have
3742 not been harvested by the Perl script yet.)
3746 Returns TRUE if the context of the currently executing subroutine is
3747 looking for a list value. Returns FALSE if the context is looking
3750 return wantarray ? () : undef;
3754 Produces a message on STDERR just like die(), but doesn't exit or throw
3757 No message is printed if there is a C<$SIG{__WARN__}> handler
3758 installed. It is the handler's responsibility to deal with the message
3759 as it sees fit (like, for instance, converting it into a die()). Most
3760 handlers must therefore make arrangements to actually display the
3761 warnings that they are not prepared to deal with, by calling warn()
3762 again in the handler. Note that this is quite safe and will not
3763 produce an endless loop, since C<__WARN__> hooks are not called from
3766 You will find this behavior is slightly different from that of
3767 C<$SIG{__DIE__}> handlers (which don't suppress the error text, but can
3768 instead call die() again to change it).
3770 Using a C<__WARN__> handler provides a powerful way to silence all
3771 warnings (even the so-called mandatory ones). An example:
3773 # wipe out *all* compile-time warnings
3774 BEGIN { $SIG{'__WARN__'} = sub { warn $_[0] if $DOWARN } }
3776 my $foo = 20; # no warning about duplicate my $foo,
3777 # but hey, you asked for it!
3778 # no compile-time or run-time warnings before here
3781 # run-time warnings enabled after here
3782 warn "\$foo is alive and $foo!"; # does show up
3784 See L<perlvar> for details on setting C<%SIG> entries, and for more
3787 =item write FILEHANDLE
3793 Writes a formatted record (possibly multi-line) to the specified file,
3794 using the format associated with that file. By default the format for
3795 a file is the one having the same name is the filehandle, but the
3796 format for the current output channel (see the select() function) may be set
3797 explicitly by assigning the name of the format to the C<$~> variable.
3799 Top of form processing is handled automatically: if there is
3800 insufficient room on the current page for the formatted record, the
3801 page is advanced by writing a form feed, a special top-of-page format
3802 is used to format the new page header, and then the record is written.
3803 By default the top-of-page format is the name of the filehandle with
3804 "_TOP" appended, but it may be dynamically set to the format of your
3805 choice by assigning the name to the C<$^> variable while the filehandle is
3806 selected. The number of lines remaining on the current page is in
3807 variable C<$->, which can be set to 0 to force a new page.
3809 If FILEHANDLE is unspecified, output goes to the current default output
3810 channel, which starts out as STDOUT but may be changed by the
3811 C<select> operator. If the FILEHANDLE is an EXPR, then the expression
3812 is evaluated and the resulting string is used to look up the name of
3813 the FILEHANDLE at run time. For more on formats, see L<perlform>.
3815 Note that write is I<NOT> the opposite of read. Unfortunately.
3819 The translation operator. See L<perlop>.