1PERLFUNC(1)            Perl Programmers Reference Guide            PERLFUNC(1)
2
3
4

NAME

6       perlfunc - Perl builtin functions
7

DESCRIPTION

9       The functions in this section can serve as terms in an expression.
10       They fall into two major categories: list operators and named unary
11       operators.  These differ in their precedence relationship with a
12       following comma.  (See the precedence table in perlop.)  List operators
13       take more than one argument, while unary operators can never take more
14       than one argument.  Thus, a comma terminates the argument of a unary
15       operator, but merely separates the arguments of a list operator.  A
16       unary operator generally provides scalar context to its argument, while
17       a list operator may provide either scalar or list contexts for its
18       arguments.  If it does both, scalar arguments come first and list
19       argument follow, and there can only ever be one such list argument.
20       For instance, "splice" has three scalar arguments followed by a list,
21       whereas "gethostbyname" has four scalar arguments.
22
23       In the syntax descriptions that follow, list operators that expect a
24       list (and provide list context for elements of the list) are shown with
25       LIST as an argument.  Such a list may consist of any combination of
26       scalar arguments or list values; the list values will be included in
27       the list as if each individual element were interpolated at that point
28       in the list, forming a longer single-dimensional list value.  Commas
29       should separate literal elements of the LIST.
30
31       Any function in the list below may be used either with or without
32       parentheses around its arguments.  (The syntax descriptions omit the
33       parentheses.)  If you use parentheses, the simple but occasionally
34       surprising rule is this: It looks like a function, therefore it is a
35       function, and precedence doesn't matter.  Otherwise it's a list
36       operator or unary operator, and precedence does matter.  Whitespace
37       between the function and left parenthesis doesn't count, so sometimes
38       you need to be careful:
39
40           print 1+2+4;      # Prints 7.
41           print(1+2) + 4;   # Prints 3.
42           print (1+2)+4;    # Also prints 3!
43           print +(1+2)+4;   # Prints 7.
44           print ((1+2)+4);  # Prints 7.
45
46       If you run Perl with the "use warnings" pragma, it can warn you about
47       this.  For example, the third line above produces:
48
49           print (...) interpreted as function at - line 1.
50           Useless use of integer addition in void context at - line 1.
51
52       A few functions take no arguments at all, and therefore work as neither
53       unary nor list operators.  These include such functions as "time" and
54       "endpwent".  For example, "time+86_400" always means "time() + 86_400".
55
56       For functions that can be used in either a scalar or list context,
57       nonabortive failure is generally indicated in scalar context by
58       returning the undefined value, and in list context by returning the
59       empty list.
60
61       Remember the following important rule: There is no rule that relates
62       the behavior of an expression in list context to its behavior in scalar
63       context, or vice versa.  It might do two totally different things.
64       Each operator and function decides which sort of value would be most
65       appropriate to return in scalar context.  Some operators return the
66       length of the list that would have been returned in list context.  Some
67       operators return the first value in the list.  Some operators return
68       the last value in the list.  Some operators return a count of
69       successful operations.  In general, they do what you want, unless you
70       want consistency.
71
72       A named array in scalar context is quite different from what would at
73       first glance appear to be a list in scalar context.  You can't get a
74       list like "(1,2,3)" into being in scalar context, because the compiler
75       knows the context at compile time.  It would generate the scalar comma
76       operator there, not the list concatenation version of the comma.  That
77       means it was never a list to start with.
78
79       In general, functions in Perl that serve as wrappers for system calls
80       ("syscalls") of the same name (like chown(2), fork(2), closedir(2),
81       etc.) return true when they succeed and "undef" otherwise, as is
82       usually mentioned in the descriptions below.  This is different from
83       the C interfaces, which return "-1" on failure.  Exceptions to this
84       rule include "wait", "waitpid", and "syscall".  System calls also set
85       the special $! variable on failure.  Other functions do not, except
86       accidentally.
87
88       Extension modules can also hook into the Perl parser to define new
89       kinds of keyword-headed expression.  These may look like functions, but
90       may also look completely different.  The syntax following the keyword
91       is defined entirely by the extension.  If you are an implementor, see
92       "PL_keyword_plugin" in perlapi for the mechanism.  If you are using
93       such a module, see the module's documentation for details of the syntax
94       that it defines.
95
96   Perl Functions by Category
97       Here are Perl's functions (including things that look like functions,
98       like some keywords and named operators) arranged by category.  Some
99       functions appear in more than one place.  Any warnings, including those
100       produced by keywords, are described in perldiag and warnings.
101
102       Functions for SCALARs or strings
103           "chomp", "chop", "chr", "crypt", "fc", "hex", "index", "lc",
104           "lcfirst", "length", "oct", "ord", "pack", "q//", "qq//",
105           "reverse", "rindex", "sprintf", "substr", "tr///", "uc", "ucfirst",
106           "y///"
107
108           "fc" is available only if the "fc" feature is enabled or if it is
109           prefixed with "CORE::".  The "fc" feature is enabled automatically
110           with a "use v5.16" (or higher) declaration in the current scope.
111
112       Regular expressions and pattern matching
113           "m//", "pos", "qr//", "quotemeta", "s///", "split", "study"
114
115       Numeric functions
116           "abs", "atan2", "cos", "exp", "hex", "int", "log", "oct", "rand",
117           "sin", "sqrt", "srand"
118
119       Functions for real @ARRAYs
120           "each", "keys", "pop", "push", "shift", "splice", "unshift",
121           "values"
122
123       Functions for list data
124           "grep", "join", "map", "qw//", "reverse", "sort", "unpack"
125
126       Functions for real %HASHes
127           "delete", "each", "exists", "keys", "values"
128
129       Input and output functions
130           "binmode", "close", "closedir", "dbmclose", "dbmopen", "die",
131           "eof", "fileno", "flock", "format", "getc", "print", "printf",
132           "read", "readdir", "readline", "rewinddir", "say", "seek",
133           "seekdir", "select", "syscall", "sysread", "sysseek", "syswrite",
134           "tell", "telldir", "truncate", "warn", "write"
135
136           "say" is available only if the "say" feature is enabled or if it is
137           prefixed with "CORE::".  The "say" feature is enabled automatically
138           with a "use v5.10" (or higher) declaration in the current scope.
139
140       Functions for fixed-length data or records
141           "pack", "read", "syscall", "sysread", "sysseek", "syswrite",
142           "unpack", "vec"
143
144       Functions for filehandles, files, or directories
145           "-X", "chdir", "chmod", "chown", "chroot", "fcntl", "glob",
146           "ioctl", "link", "lstat", "mkdir", "open", "opendir", "readlink",
147           "rename", "rmdir", "select", "stat", "symlink", "sysopen", "umask",
148           "unlink", "utime"
149
150       Keywords related to the control flow of your Perl program
151           "break", "caller", "continue", "die", "do", "dump", "eval",
152           "evalbytes", "exit", "__FILE__", "goto", "last", "__LINE__",
153           "next", "__PACKAGE__", "redo", "return", "sub", "__SUB__",
154           "wantarray"
155
156           "break" is available only if you enable the experimental "switch"
157           feature or use the "CORE::" prefix.  The "switch" feature also
158           enables the "default", "given" and "when" statements, which are
159           documented in "Switch Statements" in perlsyn.  The "switch" feature
160           is enabled automatically with a "use v5.10" (or higher) declaration
161           in the current scope.  In Perl v5.14 and earlier, "continue"
162           required the "switch" feature, like the other keywords.
163
164           "evalbytes" is only available with the "evalbytes" feature (see
165           feature) or if prefixed with "CORE::".  "__SUB__" is only available
166           with the "current_sub" feature or if prefixed with "CORE::".  Both
167           the "evalbytes" and "current_sub" features are enabled
168           automatically with a "use v5.16" (or higher) declaration in the
169           current scope.
170
171       Keywords related to scoping
172           "caller", "import", "local", "my", "our", "package", "state", "use"
173
174           "state" is available only if the "state" feature is enabled or if
175           it is prefixed with "CORE::".  The "state" feature is enabled
176           automatically with a "use v5.10" (or higher) declaration in the
177           current scope.
178
179       Miscellaneous functions
180           "defined", "formline", "lock", "prototype", "reset", "scalar",
181           "undef"
182
183       Functions for processes and process groups
184           "alarm", "exec", "fork", "getpgrp", "getppid", "getpriority",
185           "kill", "pipe", "qx//", "readpipe", "setpgrp", "setpriority",
186           "sleep", "system", "times", "wait", "waitpid"
187
188       Keywords related to Perl modules
189           "do", "import", "no", "package", "require", "use"
190
191       Keywords related to classes and object-orientation
192           "bless", "dbmclose", "dbmopen", "package", "ref", "tie", "tied",
193           "untie", "use"
194
195       Low-level socket functions
196           "accept", "bind", "connect", "getpeername", "getsockname",
197           "getsockopt", "listen", "recv", "send", "setsockopt", "shutdown",
198           "socket", "socketpair"
199
200       System V interprocess communication functions
201           "msgctl", "msgget", "msgrcv", "msgsnd", "semctl", "semget",
202           "semop", "shmctl", "shmget", "shmread", "shmwrite"
203
204       Fetching user and group info
205           "endgrent", "endhostent", "endnetent", "endpwent", "getgrent",
206           "getgrgid", "getgrnam", "getlogin", "getpwent", "getpwnam",
207           "getpwuid", "setgrent", "setpwent"
208
209       Fetching network info
210           "endprotoent", "endservent", "gethostbyaddr", "gethostbyname",
211           "gethostent", "getnetbyaddr", "getnetbyname", "getnetent",
212           "getprotobyname", "getprotobynumber", "getprotoent",
213           "getservbyname", "getservbyport", "getservent", "sethostent",
214           "setnetent", "setprotoent", "setservent"
215
216       Time-related functions
217           "gmtime", "localtime", "time", "times"
218
219       Non-function keywords
220           "and", "AUTOLOAD", "BEGIN", "CHECK", "cmp", "CORE", "__DATA__",
221           "default", "DESTROY", "else", "elseif", "elsif", "END", "__END__",
222           "eq", "for", "foreach", "ge", "given", "gt", "if", "INIT", "le",
223           "lt", "ne", "not", "or", "UNITCHECK", "unless", "until", "when",
224           "while", "x", "xor"
225
226   Portability
227       Perl was born in Unix and can therefore access all common Unix system
228       calls.  In non-Unix environments, the functionality of some Unix system
229       calls may not be available or details of the available functionality
230       may differ slightly.  The Perl functions affected by this are:
231
232       "-X", "binmode", "chmod", "chown", "chroot", "crypt", "dbmclose",
233       "dbmopen", "dump", "endgrent", "endhostent", "endnetent",
234       "endprotoent", "endpwent", "endservent", "exec", "fcntl", "flock",
235       "fork", "getgrent", "getgrgid", "gethostbyname", "gethostent",
236       "getlogin", "getnetbyaddr", "getnetbyname", "getnetent", "getppid",
237       "getpgrp", "getpriority", "getprotobynumber", "getprotoent",
238       "getpwent", "getpwnam", "getpwuid", "getservbyport", "getservent",
239       "getsockopt", "glob", "ioctl", "kill", "link", "lstat", "msgctl",
240       "msgget", "msgrcv", "msgsnd", "open", "pipe", "readlink", "rename",
241       "select", "semctl", "semget", "semop", "setgrent", "sethostent",
242       "setnetent", "setpgrp", "setpriority", "setprotoent", "setpwent",
243       "setservent", "setsockopt", "shmctl", "shmget", "shmread", "shmwrite",
244       "socket", "socketpair", "stat", "symlink", "syscall", "sysopen",
245       "system", "times", "truncate", "umask", "unlink", "utime", "wait",
246       "waitpid"
247
248       For more information about the portability of these functions, see
249       perlport and other available platform-specific documentation.
250
251   Alphabetical Listing of Perl Functions
252       -X FILEHANDLE
253       -X EXPR
254       -X DIRHANDLE
255       -X  A file test, where X is one of the letters listed below.  This
256           unary operator takes one argument, either a filename, a filehandle,
257           or a dirhandle, and tests the associated file to see if something
258           is true about it.  If the argument is omitted, tests $_, except for
259           "-t", which tests STDIN.  Unless otherwise documented, it returns 1
260           for true and '' for false.  If the file doesn't exist or can't be
261           examined, it returns "undef" and sets $! (errno).  With the
262           exception of the "-l" test they all follow symbolic links because
263           they use "stat()" and not "lstat()" (so dangling symlinks can't be
264           examined and will therefore report failure).
265
266           Despite the funny names, precedence is the same as any other named
267           unary operator.  The operator may be any of:
268
269               -r  File is readable by effective uid/gid.
270               -w  File is writable by effective uid/gid.
271               -x  File is executable by effective uid/gid.
272               -o  File is owned by effective uid.
273
274               -R  File is readable by real uid/gid.
275               -W  File is writable by real uid/gid.
276               -X  File is executable by real uid/gid.
277               -O  File is owned by real uid.
278
279               -e  File exists.
280               -z  File has zero size (is empty).
281               -s  File has nonzero size (returns size in bytes).
282
283               -f  File is a plain file.
284               -d  File is a directory.
285               -l  File is a symbolic link (false if symlinks aren't
286                   supported by the file system).
287               -p  File is a named pipe (FIFO), or Filehandle is a pipe.
288               -S  File is a socket.
289               -b  File is a block special file.
290               -c  File is a character special file.
291               -t  Filehandle is opened to a tty.
292
293               -u  File has setuid bit set.
294               -g  File has setgid bit set.
295               -k  File has sticky bit set.
296
297               -T  File is an ASCII or UTF-8 text file (heuristic guess).
298               -B  File is a "binary" file (opposite of -T).
299
300               -M  Script start time minus file modification time, in days.
301               -A  Same for access time.
302               -C  Same for inode change time (Unix, may differ for other
303                   platforms)
304
305           Example:
306
307               while (<>) {
308                   chomp;
309                   next unless -f $_;  # ignore specials
310                   #...
311               }
312
313           Note that "-s/a/b/" does not do a negated substitution.  Saying
314           "-exp($foo)" still works as expected, however: only single letters
315           following a minus are interpreted as file tests.
316
317           These operators are exempt from the "looks like a function rule"
318           described above.  That is, an opening parenthesis after the
319           operator does not affect how much of the following code constitutes
320           the argument.  Put the opening parentheses before the operator to
321           separate it from code that follows (this applies only to operators
322           with higher precedence than unary operators, of course):
323
324               -s($file) + 1024   # probably wrong; same as -s($file + 1024)
325               (-s $file) + 1024  # correct
326
327           The interpretation of the file permission operators "-r", "-R",
328           "-w", "-W", "-x", and "-X" is by default based solely on the mode
329           of the file and the uids and gids of the user.  There may be other
330           reasons you can't actually read, write, or execute the file: for
331           example network filesystem access controls, ACLs (access control
332           lists), read-only filesystems, and unrecognized executable formats.
333           Note that the use of these six specific operators to verify if some
334           operation is possible is usually a mistake, because it may be open
335           to race conditions.
336
337           Also note that, for the superuser on the local filesystems, the
338           "-r", "-R", "-w", and "-W" tests always return 1, and "-x" and "-X"
339           return 1 if any execute bit is set in the mode.  Scripts run by the
340           superuser may thus need to do a "stat" to determine the actual mode
341           of the file, or temporarily set their effective uid to something
342           else.
343
344           If you are using ACLs, there is a pragma called "filetest" that may
345           produce more accurate results than the bare "stat" mode bits.  When
346           under "use filetest 'access'", the above-mentioned filetests test
347           whether the permission can(not) be granted using the access(2)
348           family of system calls.  Also note that the "-x" and "-X" tests may
349           under this pragma return true even if there are no execute
350           permission bits set (nor any extra execute permission ACLs).  This
351           strangeness is due to the underlying system calls' definitions.
352           Note also that, due to the implementation of "use filetest
353           'access'", the "_" special filehandle won't cache the results of
354           the file tests when this pragma is in effect.  Read the
355           documentation for the "filetest" pragma for more information.
356
357           The "-T" and "-B" tests work as follows.  The first block or so of
358           the file is examined to see if it is valid UTF-8 that includes non-
359           ASCII characters.  If so, it's a "-T" file.  Otherwise, that same
360           portion of the file is examined for odd characters such as strange
361           control codes or characters with the high bit set.  If more than a
362           third of the characters are strange, it's a "-B" file; otherwise
363           it's a "-T" file.  Also, any file containing a zero byte in the
364           examined portion is considered a binary file.  (If executed within
365           the scope of a use locale which includes "LC_CTYPE", odd characters
366           are anything that isn't a printable nor space in the current
367           locale.)  If "-T" or "-B" is used on a filehandle, the current IO
368           buffer is examined rather than the first block.  Both "-T" and "-B"
369           return true on an empty file, or a file at EOF when testing a
370           filehandle.  Because you have to read a file to do the "-T" test,
371           on most occasions you want to use a "-f" against the file first, as
372           in "next unless -f $file && -T $file".
373
374           If any of the file tests (or either the "stat" or "lstat" operator)
375           is given the special filehandle consisting of a solitary underline,
376           then the stat structure of the previous file test (or "stat"
377           operator) is used, saving a system call.  (This doesn't work with
378           "-t", and you need to remember that "lstat" and "-l" leave values
379           in the stat structure for the symbolic link, not the real file.)
380           (Also, if the stat buffer was filled by an "lstat" call, "-T" and
381           "-B" will reset it with the results of "stat _").  Example:
382
383               print "Can do.\n" if -r $a || -w _ || -x _;
384
385               stat($filename);
386               print "Readable\n" if -r _;
387               print "Writable\n" if -w _;
388               print "Executable\n" if -x _;
389               print "Setuid\n" if -u _;
390               print "Setgid\n" if -g _;
391               print "Sticky\n" if -k _;
392               print "Text\n" if -T _;
393               print "Binary\n" if -B _;
394
395           As of Perl 5.10.0, as a form of purely syntactic sugar, you can
396           stack file test operators, in a way that "-f -w -x $file" is
397           equivalent to "-x $file && -w _ && -f _".  (This is only fancy
398           syntax: if you use the return value of "-f $file" as an argument to
399           another filetest operator, no special magic will happen.)
400
401           Portability issues: "-X" in perlport.
402
403           To avoid confusing would-be users of your code with mysterious
404           syntax errors, put something like this at the top of your script:
405
406               use 5.010;  # so filetest ops can stack
407
408       abs VALUE
409       abs Returns the absolute value of its argument.  If VALUE is omitted,
410           uses $_.
411
412       accept NEWSOCKET,GENERICSOCKET
413           Accepts an incoming socket connect, just as accept(2) does.
414           Returns the packed address if it succeeded, false otherwise.  See
415           the example in "Sockets: Client/Server Communication" in perlipc.
416
417           On systems that support a close-on-exec flag on files, the flag
418           will be set for the newly opened file descriptor, as determined by
419           the value of $^F.  See "$^F" in perlvar.
420
421       alarm SECONDS
422       alarm
423           Arranges to have a SIGALRM delivered to this process after the
424           specified number of wallclock seconds has elapsed.  If SECONDS is
425           not specified, the value stored in $_ is used.  (On some machines,
426           unfortunately, the elapsed time may be up to one second less or
427           more than you specified because of how seconds are counted, and
428           process scheduling may delay the delivery of the signal even
429           further.)
430
431           Only one timer may be counting at once.  Each call disables the
432           previous timer, and an argument of 0 may be supplied to cancel the
433           previous timer without starting a new one.  The returned value is
434           the amount of time remaining on the previous timer.
435
436           For delays of finer granularity than one second, the Time::HiRes
437           module (from CPAN, and starting from Perl 5.8 part of the standard
438           distribution) provides "ualarm".  You may also use Perl's four-
439           argument version of "select" leaving the first three arguments
440           undefined, or you might be able to use the "syscall" interface to
441           access setitimer(2) if your system supports it.  See perlfaq8 for
442           details.
443
444           It is usually a mistake to intermix "alarm" and "sleep" calls,
445           because "sleep" may be internally implemented on your system with
446           "alarm".
447
448           If you want to use "alarm" to time out a system call you need to
449           use an "eval"/"die" pair.  You can't rely on the alarm causing the
450           system call to fail with $! set to "EINTR" because Perl sets up
451           signal handlers to restart system calls on some systems.  Using
452           "eval"/"die" always works, modulo the caveats given in "Signals" in
453           perlipc.
454
455               eval {
456                   local $SIG{ALRM} = sub { die "alarm\n" }; # NB: \n required
457                   alarm $timeout;
458                   my $nread = sysread $socket, $buffer, $size;
459                   alarm 0;
460               };
461               if ($@) {
462                   die unless $@ eq "alarm\n";   # propagate unexpected errors
463                   # timed out
464               }
465               else {
466                   # didn't
467               }
468
469           For more information see perlipc.
470
471           Portability issues: "alarm" in perlport.
472
473       atan2 Y,X
474           Returns the arctangent of Y/X in the range -PI to PI.
475
476           For the tangent operation, you may use the "Math::Trig::tan"
477           function, or use the familiar relation:
478
479               sub tan { sin($_[0]) / cos($_[0])  }
480
481           The return value for "atan2(0,0)" is implementation-defined;
482           consult your atan2(3) manpage for more information.
483
484           Portability issues: "atan2" in perlport.
485
486       bind SOCKET,NAME
487           Binds a network address to a socket, just as bind(2) does.  Returns
488           true if it succeeded, false otherwise.  NAME should be a packed
489           address of the appropriate type for the socket.  See the examples
490           in "Sockets: Client/Server Communication" in perlipc.
491
492       binmode FILEHANDLE, LAYER
493       binmode FILEHANDLE
494           Arranges for FILEHANDLE to be read or written in "binary" or "text"
495           mode on systems where the run-time libraries distinguish between
496           binary and text files.  If FILEHANDLE is an expression, the value
497           is taken as the name of the filehandle.  Returns true on success,
498           otherwise it returns "undef" and sets $! (errno).
499
500           On some systems (in general, DOS- and Windows-based systems)
501           "binmode" is necessary when you're not working with a text file.
502           For the sake of portability it is a good idea always to use it when
503           appropriate, and never to use it when it isn't appropriate.  Also,
504           people can set their I/O to be by default UTF8-encoded Unicode, not
505           bytes.
506
507           In other words: regardless of platform, use "binmode" on binary
508           data, like images, for example.
509
510           If LAYER is present it is a single string, but may contain multiple
511           directives.  The directives alter the behaviour of the filehandle.
512           When LAYER is present, using binmode on a text file makes sense.
513
514           If LAYER is omitted or specified as ":raw" the filehandle is made
515           suitable for passing binary data.  This includes turning off
516           possible CRLF translation and marking it as bytes (as opposed to
517           Unicode characters).  Note that, despite what may be implied in
518           "Programming Perl" (the Camel, 3rd edition) or elsewhere, ":raw" is
519           not simply the inverse of ":crlf".  Other layers that would affect
520           the binary nature of the stream are also disabled.  See PerlIO, and
521           the discussion about the PERLIO environment variable in perlrun.
522
523           The ":bytes", ":crlf", ":utf8", and any other directives of the
524           form ":...", are called I/O layers.  The open pragma can be used to
525           establish default I/O layers.
526
527           The LAYER parameter of the "binmode" function is described as
528           "DISCIPLINE" in "Programming Perl, 3rd Edition".  However, since
529           the publishing of this book, by many known as "Camel III", the
530           consensus of the naming of this functionality has moved from
531           "discipline" to "layer".  All documentation of this version of Perl
532           therefore refers to "layers" rather than to "disciplines".  Now
533           back to the regularly scheduled documentation...
534
535           To mark FILEHANDLE as UTF-8, use ":utf8" or ":encoding(UTF-8)".
536           ":utf8" just marks the data as UTF-8 without further checking,
537           while ":encoding(UTF-8)" checks the data for actually being valid
538           UTF-8.  More details can be found in PerlIO::encoding.
539
540           In general, "binmode" should be called after "open" but before any
541           I/O is done on the filehandle.  Calling "binmode" normally flushes
542           any pending buffered output data (and perhaps pending input data)
543           on the handle.  An exception to this is the ":encoding" layer that
544           changes the default character encoding of the handle.  The
545           ":encoding" layer sometimes needs to be called in mid-stream, and
546           it doesn't flush the stream.  ":encoding" also implicitly pushes on
547           top of itself the ":utf8" layer because internally Perl operates on
548           UTF8-encoded Unicode characters.
549
550           The operating system, device drivers, C libraries, and Perl run-
551           time system all conspire to let the programmer treat a single
552           character ("\n") as the line terminator, irrespective of external
553           representation.  On many operating systems, the native text file
554           representation matches the internal representation, but on some
555           platforms the external representation of "\n" is made up of more
556           than one character.
557
558           All variants of Unix, Mac OS (old and new), and Stream_LF files on
559           VMS use a single character to end each line in the external
560           representation of text (even though that single character is
561           CARRIAGE RETURN on old, pre-Darwin flavors of Mac OS, and is LINE
562           FEED on Unix and most VMS files).  In other systems like OS/2, DOS,
563           and the various flavors of MS-Windows, your program sees a "\n" as
564           a simple "\cJ", but what's stored in text files are the two
565           characters "\cM\cJ".  That means that if you don't use "binmode" on
566           these systems, "\cM\cJ" sequences on disk will be converted to "\n"
567           on input, and any "\n" in your program will be converted back to
568           "\cM\cJ" on output.  This is what you want for text files, but it
569           can be disastrous for binary files.
570
571           Another consequence of using "binmode" (on some systems) is that
572           special end-of-file markers will be seen as part of the data
573           stream.  For systems from the Microsoft family this means that, if
574           your binary data contain "\cZ", the I/O subsystem will regard it as
575           the end of the file, unless you use "binmode".
576
577           "binmode" is important not only for "readline" and "print"
578           operations, but also when using "read", "seek", "sysread",
579           "syswrite" and "tell" (see perlport for more details).  See the $/
580           and "$\" variables in perlvar for how to manually set your input
581           and output line-termination sequences.
582
583           Portability issues: "binmode" in perlport.
584
585       bless REF,CLASSNAME
586       bless REF
587           This function tells the thingy referenced by REF that it is now an
588           object in the CLASSNAME package.  If CLASSNAME is an empty string,
589           it is interpreted as referring to the "main" package.  If CLASSNAME
590           is omitted, the current package is used.  Because a "bless" is
591           often the last thing in a constructor, it returns the reference for
592           convenience.  Always use the two-argument version if a derived
593           class might inherit the method doing the blessing.  See perlobj for
594           more about the blessing (and blessings) of objects.
595
596           Consider always blessing objects in CLASSNAMEs that are mixed case.
597           Namespaces with all lowercase names are considered reserved for
598           Perl pragmas.  Builtin types have all uppercase names.  To prevent
599           confusion, you may wish to avoid such package names as well.  It is
600           advised to avoid the class name 0, because much code erroneously
601           uses the result of "ref" as a truth value.
602
603           See "Perl Modules" in perlmod.
604
605       break
606           Break out of a "given" block.
607
608           "break" is available only if the "switch" feature is enabled or if
609           it is prefixed with "CORE::". The "switch" feature is enabled
610           automatically with a "use v5.10" (or higher) declaration in the
611           current scope.
612
613       caller EXPR
614       caller
615           Returns the context of the current pure perl subroutine call.  In
616           scalar context, returns the caller's package name if there is a
617           caller (that is, if we're in a subroutine or "eval" or "require")
618           and the undefined value otherwise.  caller never returns XS subs
619           and they are skipped.  The next pure perl sub will appear instead
620           of the XS sub in caller's return values.  In list context, caller
621           returns
622
623                  # 0         1          2
624               my ($package, $filename, $line) = caller;
625
626           Like "__FILE__" and "__LINE__", the filename and line number
627           returned here may be altered by the mechanism described at "Plain
628           Old Comments (Not!)" in perlsyn.
629
630           With EXPR, it returns some extra information that the debugger uses
631           to print a stack trace.  The value of EXPR indicates how many call
632           frames to go back before the current one.
633
634               #  0         1          2      3            4
635            my ($package, $filename, $line, $subroutine, $hasargs,
636
637               #  5          6          7            8       9         10
638               $wantarray, $evaltext, $is_require, $hints, $bitmask, $hinthash)
639             = caller($i);
640
641           Here, $subroutine is the function that the caller called (rather
642           than the function containing the caller).  Note that $subroutine
643           may be "(eval)" if the frame is not a subroutine call, but an
644           "eval".  In such a case additional elements $evaltext and
645           $is_require are set: $is_require is true if the frame is created by
646           a "require" or "use" statement, $evaltext contains the text of the
647           "eval EXPR" statement.  In particular, for an "eval BLOCK"
648           statement, $subroutine is "(eval)", but $evaltext is undefined.
649           (Note also that each "use" statement creates a "require" frame
650           inside an "eval EXPR" frame.)  $subroutine may also be "(unknown)"
651           if this particular subroutine happens to have been deleted from the
652           symbol table.  $hasargs is true if a new instance of @_ was set up
653           for the frame.  $hints and $bitmask contain pragmatic hints that
654           the caller was compiled with.  $hints corresponds to $^H, and
655           $bitmask corresponds to "${^WARNING_BITS}".  The $hints and
656           $bitmask values are subject to change between versions of Perl, and
657           are not meant for external use.
658
659           $hinthash is a reference to a hash containing the value of "%^H"
660           when the caller was compiled, or "undef" if "%^H" was empty.  Do
661           not modify the values of this hash, as they are the actual values
662           stored in the optree.
663
664           Furthermore, when called from within the DB package in list
665           context, and with an argument, caller returns more detailed
666           information: it sets the list variable @DB::args to be the
667           arguments with which the subroutine was invoked.
668
669           Be aware that the optimizer might have optimized call frames away
670           before "caller" had a chance to get the information.  That means
671           that caller(N) might not return information about the call frame
672           you expect it to, for "N > 1".  In particular, @DB::args might have
673           information from the previous time "caller" was called.
674
675           Be aware that setting @DB::args is best effort, intended for
676           debugging or generating backtraces, and should not be relied upon.
677           In particular, as @_ contains aliases to the caller's arguments,
678           Perl does not take a copy of @_, so @DB::args will contain
679           modifications the subroutine makes to @_ or its contents, not the
680           original values at call time.  @DB::args, like @_, does not hold
681           explicit references to its elements, so under certain cases its
682           elements may have become freed and reallocated for other variables
683           or temporary values.  Finally, a side effect of the current
684           implementation is that the effects of "shift @_" can normally be
685           undone (but not "pop @_" or other splicing, and not if a reference
686           to @_ has been taken, and subject to the caveat about reallocated
687           elements), so @DB::args is actually a hybrid of the current state
688           and initial state of @_.  Buyer beware.
689
690       chdir EXPR
691       chdir FILEHANDLE
692       chdir DIRHANDLE
693       chdir
694           Changes the working directory to EXPR, if possible.  If EXPR is
695           omitted, changes to the directory specified by $ENV{HOME}, if set;
696           if not, changes to the directory specified by $ENV{LOGDIR}.  (Under
697           VMS, the variable $ENV{'SYS$LOGIN'} is also checked, and used if it
698           is set.)  If neither is set, "chdir" does nothing and fails.  It
699           returns true on success, false otherwise.  See the example under
700           "die".
701
702           On systems that support fchdir(2), you may pass a filehandle or
703           directory handle as the argument.  On systems that don't support
704           fchdir(2), passing handles raises an exception.
705
706       chmod LIST
707           Changes the permissions of a list of files.  The first element of
708           the list must be the numeric mode, which should probably be an
709           octal number, and which definitely should not be a string of octal
710           digits: 0644 is okay, but "0644" is not.  Returns the number of
711           files successfully changed.  See also "oct" if all you have is a
712           string.
713
714               my $cnt = chmod 0755, "foo", "bar";
715               chmod 0755, @executables;
716               my $mode = "0644"; chmod $mode, "foo";      # !!! sets mode to
717                                                           # --w----r-T
718               my $mode = "0644"; chmod oct($mode), "foo"; # this is better
719               my $mode = 0644;   chmod $mode, "foo";      # this is best
720
721           On systems that support fchmod(2), you may pass filehandles among
722           the files.  On systems that don't support fchmod(2), passing
723           filehandles raises an exception.  Filehandles must be passed as
724           globs or glob references to be recognized; barewords are considered
725           filenames.
726
727               open(my $fh, "<", "foo");
728               my $perm = (stat $fh)[2] & 07777;
729               chmod($perm | 0600, $fh);
730
731           You can also import the symbolic "S_I*" constants from the "Fcntl"
732           module:
733
734               use Fcntl qw( :mode );
735               chmod S_IRWXU|S_IRGRP|S_IXGRP|S_IROTH|S_IXOTH, @executables;
736               # Identical to the chmod 0755 of the example above.
737
738           Portability issues: "chmod" in perlport.
739
740       chomp VARIABLE
741       chomp( LIST )
742       chomp
743           This safer version of "chop" removes any trailing string that
744           corresponds to the current value of $/ (also known as
745           $INPUT_RECORD_SEPARATOR in the "English" module).  It returns the
746           total number of characters removed from all its arguments.  It's
747           often used to remove the newline from the end of an input record
748           when you're worried that the final record may be missing its
749           newline.  When in paragraph mode ("$/ = ''"), it removes all
750           trailing newlines from the string.  When in slurp mode ("$/ =
751           undef") or fixed-length record mode ($/ is a reference to an
752           integer or the like; see perlvar), "chomp" won't remove anything.
753           If VARIABLE is omitted, it chomps $_.  Example:
754
755               while (<>) {
756                   chomp;  # avoid \n on last field
757                   my @array = split(/:/);
758                   # ...
759               }
760
761           If VARIABLE is a hash, it chomps the hash's values, but not its
762           keys, resetting the "each" iterator in the process.
763
764           You can actually chomp anything that's an lvalue, including an
765           assignment:
766
767               chomp(my $cwd = `pwd`);
768               chomp(my $answer = <STDIN>);
769
770           If you chomp a list, each element is chomped, and the total number
771           of characters removed is returned.
772
773           Note that parentheses are necessary when you're chomping anything
774           that is not a simple variable.  This is because "chomp $cwd =
775           `pwd`;" is interpreted as "(chomp $cwd) = `pwd`;", rather than as
776           "chomp( $cwd = `pwd` )" which you might expect.  Similarly, "chomp
777           $a, $b" is interpreted as "chomp($a), $b" rather than as "chomp($a,
778           $b)".
779
780       chop VARIABLE
781       chop( LIST )
782       chop
783           Chops off the last character of a string and returns the character
784           chopped.  It is much more efficient than "s/.$//s" because it
785           neither scans nor copies the string.  If VARIABLE is omitted, chops
786           $_.  If VARIABLE is a hash, it chops the hash's values, but not its
787           keys, resetting the "each" iterator in the process.
788
789           You can actually chop anything that's an lvalue, including an
790           assignment.
791
792           If you chop a list, each element is chopped.  Only the value of the
793           last "chop" is returned.
794
795           Note that "chop" returns the last character.  To return all but the
796           last character, use "substr($string, 0, -1)".
797
798           See also "chomp".
799
800       chown LIST
801           Changes the owner (and group) of a list of files.  The first two
802           elements of the list must be the numeric uid and gid, in that
803           order.  A value of -1 in either position is interpreted by most
804           systems to leave that value unchanged.  Returns the number of files
805           successfully changed.
806
807               my $cnt = chown $uid, $gid, 'foo', 'bar';
808               chown $uid, $gid, @filenames;
809
810           On systems that support fchown(2), you may pass filehandles among
811           the files.  On systems that don't support fchown(2), passing
812           filehandles raises an exception.  Filehandles must be passed as
813           globs or glob references to be recognized; barewords are considered
814           filenames.
815
816           Here's an example that looks up nonnumeric uids in the passwd file:
817
818               print "User: ";
819               chomp(my $user = <STDIN>);
820               print "Files: ";
821               chomp(my $pattern = <STDIN>);
822
823               my ($login,$pass,$uid,$gid) = getpwnam($user)
824                   or die "$user not in passwd file";
825
826               my @ary = glob($pattern);  # expand filenames
827               chown $uid, $gid, @ary;
828
829           On most systems, you are not allowed to change the ownership of the
830           file unless you're the superuser, although you should be able to
831           change the group to any of your secondary groups.  On insecure
832           systems, these restrictions may be relaxed, but this is not a
833           portable assumption.  On POSIX systems, you can detect this
834           condition this way:
835
836               use POSIX qw(sysconf _PC_CHOWN_RESTRICTED);
837               my $can_chown_giveaway = ! sysconf(_PC_CHOWN_RESTRICTED);
838
839           Portability issues: "chown" in perlport.
840
841       chr NUMBER
842       chr Returns the character represented by that NUMBER in the character
843           set.  For example, "chr(65)" is "A" in either ASCII or Unicode, and
844           chr(0x263a) is a Unicode smiley face.
845
846           Negative values give the Unicode replacement character
847           (chr(0xfffd)), except under the bytes pragma, where the low eight
848           bits of the value (truncated to an integer) are used.
849
850           If NUMBER is omitted, uses $_.
851
852           For the reverse, use "ord".
853
854           Note that characters from 128 to 255 (inclusive) are by default
855           internally not encoded as UTF-8 for backward compatibility reasons.
856
857           See perlunicode for more about Unicode.
858
859       chroot FILENAME
860       chroot
861           This function works like the system call by the same name: it makes
862           the named directory the new root directory for all further
863           pathnames that begin with a "/" by your process and all its
864           children.  (It doesn't change your current working directory, which
865           is unaffected.)  For security reasons, this call is restricted to
866           the superuser.  If FILENAME is omitted, does a "chroot" to $_.
867
868           NOTE:  It is mandatory for security to "chdir("/")" ("chdir" to the
869           root directory) immediately after a "chroot", otherwise the current
870           working directory may be outside of the new root.
871
872           Portability issues: "chroot" in perlport.
873
874       close FILEHANDLE
875       close
876           Closes the file or pipe associated with the filehandle, flushes the
877           IO buffers, and closes the system file descriptor.  Returns true if
878           those operations succeed and if no error was reported by any PerlIO
879           layer.  Closes the currently selected filehandle if the argument is
880           omitted.
881
882           You don't have to close FILEHANDLE if you are immediately going to
883           do another "open" on it, because "open" closes it for you.  (See
884           "open".) However, an explicit "close" on an input file resets the
885           line counter ($.), while the implicit close done by "open" does
886           not.
887
888           If the filehandle came from a piped open, "close" returns false if
889           one of the other syscalls involved fails or if its program exits
890           with non-zero status.  If the only problem was that the program
891           exited non-zero, $! will be set to 0.  Closing a pipe also waits
892           for the process executing on the pipe to exit--in case you wish to
893           look at the output of the pipe afterwards--and implicitly puts the
894           exit status value of that command into $? and
895           "${^CHILD_ERROR_NATIVE}".
896
897           If there are multiple threads running, "close" on a filehandle from
898           a piped open returns true without waiting for the child process to
899           terminate, if the filehandle is still open in another thread.
900
901           Closing the read end of a pipe before the process writing to it at
902           the other end is done writing results in the writer receiving a
903           SIGPIPE.  If the other end can't handle that, be sure to read all
904           the data before closing the pipe.
905
906           Example:
907
908               open(OUTPUT, '|sort >foo')  # pipe to sort
909                   or die "Can't start sort: $!";
910               #...                        # print stuff to output
911               close OUTPUT                # wait for sort to finish
912                   or warn $! ? "Error closing sort pipe: $!"
913                              : "Exit status $? from sort";
914               open(INPUT, 'foo')          # get sort's results
915                   or die "Can't open 'foo' for input: $!";
916
917           FILEHANDLE may be an expression whose value can be used as an
918           indirect filehandle, usually the real filehandle name or an
919           autovivified handle.
920
921       closedir DIRHANDLE
922           Closes a directory opened by "opendir" and returns the success of
923           that system call.
924
925       connect SOCKET,NAME
926           Attempts to connect to a remote socket, just like connect(2).
927           Returns true if it succeeded, false otherwise.  NAME should be a
928           packed address of the appropriate type for the socket.  See the
929           examples in "Sockets: Client/Server Communication" in perlipc.
930
931       continue BLOCK
932       continue
933           When followed by a BLOCK, "continue" is actually a flow control
934           statement rather than a function.  If there is a "continue" BLOCK
935           attached to a BLOCK (typically in a "while" or "foreach"), it is
936           always executed just before the conditional is about to be
937           evaluated again, just like the third part of a "for" loop in C.
938           Thus it can be used to increment a loop variable, even when the
939           loop has been continued via the "next" statement (which is similar
940           to the C "continue" statement).
941
942           "last", "next", or "redo" may appear within a "continue" block;
943           "last" and "redo" behave as if they had been executed within the
944           main block.  So will "next", but since it will execute a "continue"
945           block, it may be more entertaining.
946
947               while (EXPR) {
948                   ### redo always comes here
949                   do_something;
950               } continue {
951                   ### next always comes here
952                   do_something_else;
953                   # then back the top to re-check EXPR
954               }
955               ### last always comes here
956
957           Omitting the "continue" section is equivalent to using an empty
958           one, logically enough, so "next" goes directly back to check the
959           condition at the top of the loop.
960
961           When there is no BLOCK, "continue" is a function that falls through
962           the current "when" or "default" block instead of iterating a
963           dynamically enclosing "foreach" or exiting a lexically enclosing
964           "given".  In Perl 5.14 and earlier, this form of "continue" was
965           only available when the "switch" feature was enabled.  See feature
966           and "Switch Statements" in perlsyn for more information.
967
968       cos EXPR
969       cos Returns the cosine of EXPR (expressed in radians).  If EXPR is
970           omitted, takes the cosine of $_.
971
972           For the inverse cosine operation, you may use the
973           "Math::Trig::acos" function, or use this relation:
974
975               sub acos { atan2( sqrt(1 - $_[0] * $_[0]), $_[0] ) }
976
977       crypt PLAINTEXT,SALT
978           Creates a digest string exactly like the crypt(3) function in the C
979           library (assuming that you actually have a version there that has
980           not been extirpated as a potential munition).
981
982           "crypt" is a one-way hash function.  The PLAINTEXT and SALT are
983           turned into a short string, called a digest, which is returned.
984           The same PLAINTEXT and SALT will always return the same string, but
985           there is no (known) way to get the original PLAINTEXT from the
986           hash.  Small changes in the PLAINTEXT or SALT will result in large
987           changes in the digest.
988
989           There is no decrypt function.  This function isn't all that useful
990           for cryptography (for that, look for Crypt modules on your nearby
991           CPAN mirror) and the name "crypt" is a bit of a misnomer.  Instead
992           it is primarily used to check if two pieces of text are the same
993           without having to transmit or store the text itself.  An example is
994           checking if a correct password is given.  The digest of the
995           password is stored, not the password itself.  The user types in a
996           password that is "crypt"'d with the same salt as the stored digest.
997           If the two digests match, the password is correct.
998
999           When verifying an existing digest string you should use the digest
1000           as the salt (like "crypt($plain, $digest) eq $digest").  The SALT
1001           used to create the digest is visible as part of the digest.  This
1002           ensures "crypt" will hash the new string with the same salt as the
1003           digest.  This allows your code to work with the standard "crypt"
1004           and with more exotic implementations.  In other words, assume
1005           nothing about the returned string itself nor about how many bytes
1006           of SALT may matter.
1007
1008           Traditionally the result is a string of 13 bytes: two first bytes
1009           of the salt, followed by 11 bytes from the set "[./0-9A-Za-z]", and
1010           only the first eight bytes of PLAINTEXT mattered.  But alternative
1011           hashing schemes (like MD5), higher level security schemes (like
1012           C2), and implementations on non-Unix platforms may produce
1013           different strings.
1014
1015           When choosing a new salt create a random two character string whose
1016           characters come from the set "[./0-9A-Za-z]" (like "join '', ('.',
1017           '/', 0..9, 'A'..'Z', 'a'..'z')[rand 64, rand 64]").  This set of
1018           characters is just a recommendation; the characters allowed in the
1019           salt depend solely on your system's crypt library, and Perl can't
1020           restrict what salts "crypt" accepts.
1021
1022           Here's an example that makes sure that whoever runs this program
1023           knows their password:
1024
1025               my $pwd = (getpwuid($<))[1];
1026
1027               system "stty -echo";
1028               print "Password: ";
1029               chomp(my $word = <STDIN>);
1030               print "\n";
1031               system "stty echo";
1032
1033               if (crypt($word, $pwd) ne $pwd) {
1034                   die "Sorry...\n";
1035               } else {
1036                   print "ok\n";
1037               }
1038
1039           Of course, typing in your own password to whoever asks you for it
1040           is unwise.
1041
1042           The "crypt" function is unsuitable for hashing large quantities of
1043           data, not least of all because you can't get the information back.
1044           Look at the Digest module for more robust algorithms.
1045
1046           If using "crypt" on a Unicode string (which potentially has
1047           characters with codepoints above 255), Perl tries to make sense of
1048           the situation by trying to downgrade (a copy of) the string back to
1049           an eight-bit byte string before calling "crypt" (on that copy).  If
1050           that works, good.  If not, "crypt" dies with "Wide character in
1051           crypt".
1052
1053           Portability issues: "crypt" in perlport.
1054
1055       dbmclose HASH
1056           [This function has been largely superseded by the "untie"
1057           function.]
1058
1059           Breaks the binding between a DBM file and a hash.
1060
1061           Portability issues: "dbmclose" in perlport.
1062
1063       dbmopen HASH,DBNAME,MASK
1064           [This function has been largely superseded by the "tie" function.]
1065
1066           This binds a dbm(3), ndbm(3), sdbm(3), gdbm(3), or Berkeley DB file
1067           to a hash.  HASH is the name of the hash.  (Unlike normal "open",
1068           the first argument is not a filehandle, even though it looks like
1069           one).  DBNAME is the name of the database (without the .dir or .pag
1070           extension if any).  If the database does not exist, it is created
1071           with protection specified by MASK (as modified by the "umask").  To
1072           prevent creation of the database if it doesn't exist, you may
1073           specify a MODE of 0, and the function will return a false value if
1074           it can't find an existing database.  If your system supports only
1075           the older DBM functions, you may make only one "dbmopen" call in
1076           your program.  In older versions of Perl, if your system had
1077           neither DBM nor ndbm, calling "dbmopen" produced a fatal error; it
1078           now falls back to sdbm(3).
1079
1080           If you don't have write access to the DBM file, you can only read
1081           hash variables, not set them.  If you want to test whether you can
1082           write, either use file tests or try setting a dummy hash entry
1083           inside an "eval" to trap the error.
1084
1085           Note that functions such as "keys" and "values" may return huge
1086           lists when used on large DBM files.  You may prefer to use the
1087           "each" function to iterate over large DBM files.  Example:
1088
1089               # print out history file offsets
1090               dbmopen(%HIST,'/usr/lib/news/history',0666);
1091               while (($key,$val) = each %HIST) {
1092                   print $key, ' = ', unpack('L',$val), "\n";
1093               }
1094               dbmclose(%HIST);
1095
1096           See also AnyDBM_File for a more general description of the pros and
1097           cons of the various dbm approaches, as well as DB_File for a
1098           particularly rich implementation.
1099
1100           You can control which DBM library you use by loading that library
1101           before you call "dbmopen":
1102
1103               use DB_File;
1104               dbmopen(%NS_Hist, "$ENV{HOME}/.netscape/history.db")
1105                   or die "Can't open netscape history file: $!";
1106
1107           Portability issues: "dbmopen" in perlport.
1108
1109       defined EXPR
1110       defined
1111           Returns a Boolean value telling whether EXPR has a value other than
1112           the undefined value "undef".  If EXPR is not present, $_ is
1113           checked.
1114
1115           Many operations return "undef" to indicate failure, end of file,
1116           system error, uninitialized variable, and other exceptional
1117           conditions.  This function allows you to distinguish "undef" from
1118           other values.  (A simple Boolean test will not distinguish among
1119           "undef", zero, the empty string, and "0", which are all equally
1120           false.)  Note that since "undef" is a valid scalar, its presence
1121           doesn't necessarily indicate an exceptional condition: "pop"
1122           returns "undef" when its argument is an empty array, or when the
1123           element to return happens to be "undef".
1124
1125           You may also use "defined(&func)" to check whether subroutine
1126           "func" has ever been defined.  The return value is unaffected by
1127           any forward declarations of "func".  A subroutine that is not
1128           defined may still be callable: its package may have an "AUTOLOAD"
1129           method that makes it spring into existence the first time that it
1130           is called; see perlsub.
1131
1132           Use of "defined" on aggregates (hashes and arrays) is no longer
1133           supported. It used to report whether memory for that aggregate had
1134           ever been allocated.  You should instead use a simple test for
1135           size:
1136
1137               if (@an_array) { print "has array elements\n" }
1138               if (%a_hash)   { print "has hash members\n"   }
1139
1140           When used on a hash element, it tells you whether the value is
1141           defined, not whether the key exists in the hash.  Use "exists" for
1142           the latter purpose.
1143
1144           Examples:
1145
1146               print if defined $switch{D};
1147               print "$val\n" while defined($val = pop(@ary));
1148               die "Can't readlink $sym: $!"
1149                   unless defined($value = readlink $sym);
1150               sub foo { defined &$bar ? $bar->(@_) : die "No bar"; }
1151               $debugging = 0 unless defined $debugging;
1152
1153           Note:  Many folks tend to overuse "defined" and are then surprised
1154           to discover that the number 0 and "" (the zero-length string) are,
1155           in fact, defined values.  For example, if you say
1156
1157               "ab" =~ /a(.*)b/;
1158
1159           The pattern match succeeds and $1 is defined, although it matched
1160           "nothing".  It didn't really fail to match anything.  Rather, it
1161           matched something that happened to be zero characters long.  This
1162           is all very above-board and honest.  When a function returns an
1163           undefined value, it's an admission that it couldn't give you an
1164           honest answer.  So you should use "defined" only when questioning
1165           the integrity of what you're trying to do.  At other times, a
1166           simple comparison to 0 or "" is what you want.
1167
1168           See also "undef", "exists", "ref".
1169
1170       delete EXPR
1171           Given an expression that specifies an element or slice of a hash,
1172           "delete" deletes the specified elements from that hash so that
1173           "exists" on that element no longer returns true.  Setting a hash
1174           element to the undefined value does not remove its key, but
1175           deleting it does; see "exists".
1176
1177           In list context, usually returns the value or values deleted, or
1178           the last such element in scalar context.  The return list's length
1179           corresponds to that of the argument list: deleting non-existent
1180           elements returns the undefined value in their corresponding
1181           positions. When a key/value hash slice is passed to "delete", the
1182           return value is a list of key/value pairs (two elements for each
1183           item deleted from the hash).
1184
1185           "delete" may also be used on arrays and array slices, but its
1186           behavior is less straightforward.  Although "exists" will return
1187           false for deleted entries, deleting array elements never changes
1188           indices of existing values; use "shift" or "splice" for that.
1189           However, if any deleted elements fall at the end of an array, the
1190           array's size shrinks to the position of the highest element that
1191           still tests true for "exists", or to 0 if none do.  In other words,
1192           an array won't have trailing nonexistent elements after a delete.
1193
1194           WARNING: Calling "delete" on array values is strongly discouraged.
1195           The notion of deleting or checking the existence of Perl array
1196           elements is not conceptually coherent, and can lead to surprising
1197           behavior.
1198
1199           Deleting from %ENV modifies the environment.  Deleting from a hash
1200           tied to a DBM file deletes the entry from the DBM file.  Deleting
1201           from a "tied" hash or array may not necessarily return anything; it
1202           depends on the implementation of the "tied" package's DELETE
1203           method, which may do whatever it pleases.
1204
1205           The "delete local EXPR" construct localizes the deletion to the
1206           current block at run time.  Until the block exits, elements locally
1207           deleted temporarily no longer exist.  See "Localized deletion of
1208           elements of composite types" in perlsub.
1209
1210               my %hash = (foo => 11, bar => 22, baz => 33);
1211               my $scalar = delete $hash{foo};         # $scalar is 11
1212               $scalar = delete @hash{qw(foo bar)}; # $scalar is 22
1213               my @array  = delete @hash{qw(foo baz)}; # @array  is (undef,33)
1214
1215           The following (inefficiently) deletes all the values of %HASH and
1216           @ARRAY:
1217
1218               foreach my $key (keys %HASH) {
1219                   delete $HASH{$key};
1220               }
1221
1222               foreach my $index (0 .. $#ARRAY) {
1223                   delete $ARRAY[$index];
1224               }
1225
1226           And so do these:
1227
1228               delete @HASH{keys %HASH};
1229
1230               delete @ARRAY[0 .. $#ARRAY];
1231
1232           But both are slower than assigning the empty list or undefining
1233           %HASH or @ARRAY, which is the customary way to empty out an
1234           aggregate:
1235
1236               %HASH = ();     # completely empty %HASH
1237               undef %HASH;    # forget %HASH ever existed
1238
1239               @ARRAY = ();    # completely empty @ARRAY
1240               undef @ARRAY;   # forget @ARRAY ever existed
1241
1242           The EXPR can be arbitrarily complicated provided its final
1243           operation is an element or slice of an aggregate:
1244
1245               delete $ref->[$x][$y]{$key};
1246               delete @{$ref->[$x][$y]}{$key1, $key2, @morekeys};
1247
1248               delete $ref->[$x][$y][$index];
1249               delete @{$ref->[$x][$y]}[$index1, $index2, @moreindices];
1250
1251       die LIST
1252           "die" raises an exception.  Inside an "eval" the exception is
1253           stuffed into $@ and the "eval" is terminated with the undefined
1254           value.  If the exception is outside of all enclosing "eval"s, then
1255           the uncaught exception is printed to "STDERR" and perl exits with
1256           an exit code indicating failure.  If you need to exit the process
1257           with a specific exit code, see "exit".
1258
1259           Equivalent examples:
1260
1261               die "Can't cd to spool: $!\n" unless chdir '/usr/spool/news';
1262               chdir '/usr/spool/news' or die "Can't cd to spool: $!\n"
1263
1264           Most of the time, "die" is called with a string to use as the
1265           exception.  You may either give a single non-reference operand to
1266           serve as the exception, or a list of two or more items, which will
1267           be stringified and concatenated to make the exception.
1268
1269           If the string exception does not end in a newline, the current
1270           script line number and input line number (if any) and a newline are
1271           appended to it.  Note that the "input line number" (also known as
1272           "chunk") is subject to whatever notion of "line" happens to be
1273           currently in effect, and is also available as the special variable
1274           $..  See "$/" in perlvar and "$." in perlvar.
1275
1276           Hint: sometimes appending ", stopped" to your message will cause it
1277           to make better sense when the string "at foo line 123" is appended.
1278           Suppose you are running script "canasta".
1279
1280               die "/etc/games is no good";
1281               die "/etc/games is no good, stopped";
1282
1283           produce, respectively
1284
1285               /etc/games is no good at canasta line 123.
1286               /etc/games is no good, stopped at canasta line 123.
1287
1288           If LIST was empty or made an empty string, and $@ already contains
1289           an exception value (typically from a previous "eval"), then that
1290           value is reused after appending "\t...propagated".  This is useful
1291           for propagating exceptions:
1292
1293               eval { ... };
1294               die unless $@ =~ /Expected exception/;
1295
1296           If LIST was empty or made an empty string, and $@ contains an
1297           object reference that has a "PROPAGATE" method, that method will be
1298           called with additional file and line number parameters.  The return
1299           value replaces the value in $@;  i.e., as if "$@ = eval {
1300           $@->PROPAGATE(__FILE__, __LINE__) };" were called.
1301
1302           If LIST was empty or made an empty string, and $@ is also empty,
1303           then the string "Died" is used.
1304
1305           You can also call "die" with a reference argument, and if this is
1306           trapped within an "eval", $@ contains that reference.  This permits
1307           more elaborate exception handling using objects that maintain
1308           arbitrary state about the exception.  Such a scheme is sometimes
1309           preferable to matching particular string values of $@ with regular
1310           expressions.
1311
1312           Because Perl stringifies uncaught exception messages before
1313           display, you'll probably want to overload stringification
1314           operations on exception objects.  See overload for details about
1315           that.  The stringified message should be non-empty, and should end
1316           in a newline, in order to fit in with the treatment of string
1317           exceptions.  Also, because an exception object reference cannot be
1318           stringified without destroying it, Perl doesn't attempt to append
1319           location or other information to a reference exception.  If you
1320           want location information with a complex exception object, you'll
1321           have to arrange to put the location information into the object
1322           yourself.
1323
1324           Because $@ is a global variable, be careful that analyzing an
1325           exception caught by "eval" doesn't replace the reference in the
1326           global variable.  It's easiest to make a local copy of the
1327           reference before any manipulations.  Here's an example:
1328
1329               use Scalar::Util "blessed";
1330
1331               eval { ... ; die Some::Module::Exception->new( FOO => "bar" ) };
1332               if (my $ev_err = $@) {
1333                   if (blessed($ev_err)
1334                       && $ev_err->isa("Some::Module::Exception")) {
1335                       # handle Some::Module::Exception
1336                   }
1337                   else {
1338                       # handle all other possible exceptions
1339                   }
1340               }
1341
1342           If an uncaught exception results in interpreter exit, the exit code
1343           is determined from the values of $! and $? with this pseudocode:
1344
1345               exit $! if $!;              # errno
1346               exit $? >> 8 if $? >> 8;    # child exit status
1347               exit 255;                   # last resort
1348
1349           As with "exit", $? is set prior to unwinding the call stack; any
1350           "DESTROY" or "END" handlers can then alter this value, and thus
1351           Perl's exit code.
1352
1353           The intent is to squeeze as much possible information about the
1354           likely cause into the limited space of the system exit code.
1355           However, as $! is the value of C's "errno", which can be set by any
1356           system call, this means that the value of the exit code used by
1357           "die" can be non-predictable, so should not be relied upon, other
1358           than to be non-zero.
1359
1360           You can arrange for a callback to be run just before the "die" does
1361           its deed, by setting the $SIG{__DIE__} hook.  The associated
1362           handler is called with the exception as an argument, and can change
1363           the exception, if it sees fit, by calling "die" again.  See "%SIG"
1364           in perlvar for details on setting %SIG entries, and "eval" for some
1365           examples.  Although this feature was to be run only right before
1366           your program was to exit, this is not currently so: the
1367           $SIG{__DIE__} hook is currently called even inside "eval"ed
1368           blocks/strings!  If one wants the hook to do nothing in such
1369           situations, put
1370
1371               die @_ if $^S;
1372
1373           as the first line of the handler (see "$^S" in perlvar).  Because
1374           this promotes strange action at a distance, this counterintuitive
1375           behavior may be fixed in a future release.
1376
1377           See also "exit", "warn", and the Carp module.
1378
1379       do BLOCK
1380           Not really a function.  Returns the value of the last command in
1381           the sequence of commands indicated by BLOCK.  When modified by the
1382           "while" or "until" loop modifier, executes the BLOCK once before
1383           testing the loop condition.  (On other statements the loop
1384           modifiers test the conditional first.)
1385
1386           "do BLOCK" does not count as a loop, so the loop control statements
1387           "next", "last", or "redo" cannot be used to leave or restart the
1388           block.  See perlsyn for alternative strategies.
1389
1390       do EXPR
1391           Uses the value of EXPR as a filename and executes the contents of
1392           the file as a Perl script:
1393
1394               # load the exact specified file (./ and ../ special-cased)
1395               do '/foo/stat.pl';
1396               do './stat.pl';
1397               do '../foo/stat.pl';
1398
1399               # search for the named file within @INC
1400               do 'stat.pl';
1401               do 'foo/stat.pl';
1402
1403           "do './stat.pl'" is largely like
1404
1405               eval `cat stat.pl`;
1406
1407           except that it's more concise, runs no external processes, and
1408           keeps track of the current filename for error messages. It also
1409           differs in that code evaluated with "do FILE" cannot see lexicals
1410           in the enclosing scope; "eval STRING" does.  It's the same,
1411           however, in that it does reparse the file every time you call it,
1412           so you probably don't want to do this inside a loop.
1413
1414           Using "do" with a relative path (except for ./ and ../), like
1415
1416               do 'foo/stat.pl';
1417
1418           will search the @INC directories, and update %INC if the file is
1419           found.  See "@INC" in perlvar and "%INC" in perlvar for these
1420           variables. In particular, note that whilst historically @INC
1421           contained '.' (the current directory) making these two cases
1422           equivalent, that is no longer necessarily the case, as '.' is not
1423           included in @INC by default in perl versions 5.26.0 onwards.
1424           Instead, perl will now warn:
1425
1426               do "stat.pl" failed, '.' is no longer in @INC;
1427               did you mean do "./stat.pl"?
1428
1429           If "do" can read the file but cannot compile it, it returns "undef"
1430           and sets an error message in $@.  If "do" cannot read the file, it
1431           returns undef and sets $! to the error.  Always check $@ first, as
1432           compilation could fail in a way that also sets $!.  If the file is
1433           successfully compiled, "do" returns the value of the last
1434           expression evaluated.
1435
1436           Inclusion of library modules is better done with the "use" and
1437           "require" operators, which also do automatic error checking and
1438           raise an exception if there's a problem.
1439
1440           You might like to use "do" to read in a program configuration file.
1441           Manual error checking can be done this way:
1442
1443               # Read in config files: system first, then user.
1444               # Beware of using relative pathnames here.
1445               for $file ("/share/prog/defaults.rc",
1446                          "$ENV{HOME}/.someprogrc")
1447               {
1448                   unless ($return = do $file) {
1449                       warn "couldn't parse $file: $@" if $@;
1450                       warn "couldn't do $file: $!"    unless defined $return;
1451                       warn "couldn't run $file"       unless $return;
1452                   }
1453               }
1454
1455       dump LABEL
1456       dump EXPR
1457       dump
1458           This function causes an immediate core dump.  See also the -u
1459           command-line switch in perlrun, which does the same thing.
1460           Primarily this is so that you can use the undump program (not
1461           supplied) to turn your core dump into an executable binary after
1462           having initialized all your variables at the beginning of the
1463           program.  When the new binary is executed it will begin by
1464           executing a "goto LABEL" (with all the restrictions that "goto"
1465           suffers).  Think of it as a goto with an intervening core dump and
1466           reincarnation.  If "LABEL" is omitted, restarts the program from
1467           the top.  The "dump EXPR" form, available starting in Perl 5.18.0,
1468           allows a name to be computed at run time, being otherwise identical
1469           to "dump LABEL".
1470
1471           WARNING: Any files opened at the time of the dump will not be open
1472           any more when the program is reincarnated, with possible resulting
1473           confusion by Perl.
1474
1475           This function is now largely obsolete, mostly because it's very
1476           hard to convert a core file into an executable.  As of Perl 5.30,
1477           it must be invoked as "CORE::dump()".
1478
1479           Unlike most named operators, this has the same precedence as
1480           assignment.  It is also exempt from the looks-like-a-function rule,
1481           so "dump ("foo")."bar"" will cause "bar" to be part of the argument
1482           to "dump".
1483
1484           Portability issues: "dump" in perlport.
1485
1486       each HASH
1487       each ARRAY
1488           When called on a hash in list context, returns a 2-element list
1489           consisting of the key and value for the next element of a hash.  In
1490           Perl 5.12 and later only, it will also return the index and value
1491           for the next element of an array so that you can iterate over it;
1492           older Perls consider this a syntax error.  When called in scalar
1493           context, returns only the key (not the value) in a hash, or the
1494           index in an array.
1495
1496           Hash entries are returned in an apparently random order.  The
1497           actual random order is specific to a given hash; the exact same
1498           series of operations on two hashes may result in a different order
1499           for each hash.  Any insertion into the hash may change the order,
1500           as will any deletion, with the exception that the most recent key
1501           returned by "each" or "keys" may be deleted without changing the
1502           order.  So long as a given hash is unmodified you may rely on
1503           "keys", "values" and "each" to repeatedly return the same order as
1504           each other.  See "Algorithmic Complexity Attacks" in perlsec for
1505           details on why hash order is randomized.  Aside from the guarantees
1506           provided here the exact details of Perl's hash algorithm and the
1507           hash traversal order are subject to change in any release of Perl.
1508
1509           After "each" has returned all entries from the hash or array, the
1510           next call to "each" returns the empty list in list context and
1511           "undef" in scalar context; the next call following that one
1512           restarts iteration.  Each hash or array has its own internal
1513           iterator, accessed by "each", "keys", and "values".  The iterator
1514           is implicitly reset when "each" has reached the end as just
1515           described; it can be explicitly reset by calling "keys" or "values"
1516           on the hash or array, or by referencing the hash (but not array) in
1517           list context.  If you add or delete a hash's elements while
1518           iterating over it, the effect on the iterator is unspecified; for
1519           example, entries may be skipped or duplicated--so don't do that.
1520           Exception: It is always safe to delete the item most recently
1521           returned by "each", so the following code works properly:
1522
1523               while (my ($key, $value) = each %hash) {
1524                   print $key, "\n";
1525                   delete $hash{$key};   # This is safe
1526               }
1527
1528           Tied hashes may have a different ordering behaviour to perl's hash
1529           implementation.
1530
1531           The iterator used by "each" is attached to the hash or array, and
1532           is shared between all iteration operations applied to the same hash
1533           or array.  Thus all uses of "each" on a single hash or array
1534           advance the same iterator location.  All uses of "each" are also
1535           subject to having the iterator reset by any use of "keys" or
1536           "values" on the same hash or array, or by the hash (but not array)
1537           being referenced in list context.  This makes "each"-based loops
1538           quite fragile: it is easy to arrive at such a loop with the
1539           iterator already part way through the object, or to accidentally
1540           clobber the iterator state during execution of the loop body.  It's
1541           easy enough to explicitly reset the iterator before starting a
1542           loop, but there is no way to insulate the iterator state used by a
1543           loop from the iterator state used by anything else that might
1544           execute during the loop body.  To avoid these problems, use a
1545           "foreach" loop rather than "while"-"each".
1546
1547           This prints out your environment like the printenv(1) program, but
1548           in a different order:
1549
1550               while (my ($key,$value) = each %ENV) {
1551                   print "$key=$value\n";
1552               }
1553
1554           Starting with Perl 5.14, an experimental feature allowed "each" to
1555           take a scalar expression. This experiment has been deemed
1556           unsuccessful, and was removed as of Perl 5.24.
1557
1558           As of Perl 5.18 you can use a bare "each" in a "while" loop, which
1559           will set $_ on every iteration.  If either an "each" expression or
1560           an explicit assignment of an "each" expression to a scalar is used
1561           as a "while"/"for" condition, then the condition actually tests for
1562           definedness of the expression's value, not for its regular truth
1563           value.
1564
1565               while (each %ENV) {
1566                   print "$_=$ENV{$_}\n";
1567               }
1568
1569           To avoid confusing would-be users of your code who are running
1570           earlier versions of Perl with mysterious syntax errors, put this
1571           sort of thing at the top of your file to signal that your code will
1572           work only on Perls of a recent vintage:
1573
1574               use 5.012;  # so keys/values/each work on arrays
1575               use 5.018;  # so each assigns to $_ in a lone while test
1576
1577           See also "keys", "values", and "sort".
1578
1579       eof FILEHANDLE
1580       eof ()
1581       eof Returns 1 if the next read on FILEHANDLE will return end of file or
1582           if FILEHANDLE is not open.  FILEHANDLE may be an expression whose
1583           value gives the real filehandle.  (Note that this function actually
1584           reads a character and then "ungetc"s it, so isn't useful in an
1585           interactive context.)  Do not read from a terminal file (or call
1586           "eof(FILEHANDLE)" on it) after end-of-file is reached.  File types
1587           such as terminals may lose the end-of-file condition if you do.
1588
1589           An "eof" without an argument uses the last file read.  Using
1590           "eof()" with empty parentheses is different.  It refers to the
1591           pseudo file formed from the files listed on the command line and
1592           accessed via the "<>" operator.  Since "<>" isn't explicitly
1593           opened, as a normal filehandle is, an "eof()" before "<>" has been
1594           used will cause @ARGV to be examined to determine if input is
1595           available.   Similarly, an "eof()" after "<>" has returned end-of-
1596           file will assume you are processing another @ARGV list, and if you
1597           haven't set @ARGV, will read input from "STDIN"; see "I/O
1598           Operators" in perlop.
1599
1600           In a "while (<>)" loop, "eof" or "eof(ARGV)" can be used to detect
1601           the end of each file, whereas "eof()" will detect the end of the
1602           very last file only.  Examples:
1603
1604               # reset line numbering on each input file
1605               while (<>) {
1606                   next if /^\s*#/;  # skip comments
1607                   print "$.\t$_";
1608               } continue {
1609                   close ARGV if eof;  # Not eof()!
1610               }
1611
1612               # insert dashes just before last line of last file
1613               while (<>) {
1614                   if (eof()) {  # check for end of last file
1615                       print "--------------\n";
1616                   }
1617                   print;
1618                   last if eof();     # needed if we're reading from a terminal
1619               }
1620
1621           Practical hint: you almost never need to use "eof" in Perl, because
1622           the input operators typically return "undef" when they run out of
1623           data or encounter an error.
1624
1625       eval EXPR
1626       eval BLOCK
1627       eval
1628           "eval" in all its forms is used to execute a little Perl program,
1629           trapping any errors encountered so they don't crash the calling
1630           program.
1631
1632           Plain "eval" with no argument is just "eval EXPR", where the
1633           expression is understood to be contained in $_.  Thus there are
1634           only two real "eval" forms; the one with an EXPR is often called
1635           "string eval".  In a string eval, the value of the expression
1636           (which is itself determined within scalar context) is first parsed,
1637           and if there were no errors, executed as a block within the lexical
1638           context of the current Perl program.  This form is typically used
1639           to delay parsing and subsequent execution of the text of EXPR until
1640           run time.  Note that the value is parsed every time the "eval"
1641           executes.
1642
1643           The other form is called "block eval".  It is less general than
1644           string eval, but the code within the BLOCK is parsed only once (at
1645           the same time the code surrounding the "eval" itself was parsed)
1646           and executed within the context of the current Perl program.  This
1647           form is typically used to trap exceptions more efficiently than the
1648           first, while also providing the benefit of checking the code within
1649           BLOCK at compile time.  BLOCK is parsed and compiled just once.
1650           Since errors are trapped, it often is used to check if a given
1651           feature is available.
1652
1653           In both forms, the value returned is the value of the last
1654           expression evaluated inside the mini-program; a return statement
1655           may also be used, just as with subroutines.  The expression
1656           providing the return value is evaluated in void, scalar, or list
1657           context, depending on the context of the "eval" itself.  See
1658           "wantarray" for more on how the evaluation context can be
1659           determined.
1660
1661           If there is a syntax error or runtime error, or a "die" statement
1662           is executed, "eval" returns "undef" in scalar context, or an empty
1663           list in list context, and $@ is set to the error message.  (Prior
1664           to 5.16, a bug caused "undef" to be returned in list context for
1665           syntax errors, but not for runtime errors.) If there was no error,
1666           $@ is set to the empty string.  A control flow operator like "last"
1667           or "goto" can bypass the setting of $@.  Beware that using "eval"
1668           neither silences Perl from printing warnings to STDERR, nor does it
1669           stuff the text of warning messages into $@.  To do either of those,
1670           you have to use the $SIG{__WARN__} facility, or turn off warnings
1671           inside the BLOCK or EXPR using "no warnings 'all'".  See "warn",
1672           perlvar, and warnings.
1673
1674           Note that, because "eval" traps otherwise-fatal errors, it is
1675           useful for determining whether a particular feature (such as
1676           "socket" or "symlink") is implemented.  It is also Perl's
1677           exception-trapping mechanism, where the "die" operator is used to
1678           raise exceptions.
1679
1680           Before Perl 5.14, the assignment to $@ occurred before restoration
1681           of localized variables, which means that for your code to run on
1682           older versions, a temporary is required if you want to mask some,
1683           but not all errors:
1684
1685            # alter $@ on nefarious repugnancy only
1686            {
1687               my $e;
1688               {
1689                 local $@; # protect existing $@
1690                 eval { test_repugnancy() };
1691                 # $@ =~ /nefarious/ and die $@; # Perl 5.14 and higher only
1692                 $@ =~ /nefarious/ and $e = $@;
1693               }
1694               die $e if defined $e
1695            }
1696
1697           There are some different considerations for each form:
1698
1699           String eval
1700               Since the return value of EXPR is executed as a block within
1701               the lexical context of the current Perl program, any outer
1702               lexical variables are visible to it, and any package variable
1703               settings or subroutine and format definitions remain
1704               afterwards.
1705
1706               Under the "unicode_eval" feature
1707                   If this feature is enabled (which is the default under a
1708                   "use 5.16" or higher declaration), EXPR is considered to be
1709                   in the same encoding as the surrounding program.  Thus if
1710                   "use utf8" is in effect, the string will be treated as
1711                   being UTF-8 encoded.  Otherwise, the string is considered
1712                   to be a sequence of independent bytes.  Bytes that
1713                   correspond to ASCII-range code points will have their
1714                   normal meanings for operators in the string.  The treatment
1715                   of the other bytes depends on if the "'unicode_strings""
1716                   feature is in effect.
1717
1718                   In a plain "eval" without an EXPR argument, being in
1719                   "use utf8" or not is irrelevant; the UTF-8ness of $_ itself
1720                   determines the behavior.
1721
1722                   Any "use utf8" or "no utf8" declarations within the string
1723                   have no effect, and source filters are forbidden.
1724                   ("unicode_strings", however, can appear within the string.)
1725                   See also the "evalbytes" operator, which works properly
1726                   with source filters.
1727
1728                   Variables defined outside the "eval" and used inside it
1729                   retain their original UTF-8ness.  Everything inside the
1730                   string follows the normal rules for a Perl program with the
1731                   given state of "use utf8".
1732
1733               Outside the "unicode_eval" feature
1734                   In this case, the behavior is problematic and is not so
1735                   easily described.  Here are two bugs that cannot easily be
1736                   fixed without breaking existing programs:
1737
1738                   •   It can lose track of whether something should be
1739                       encoded as UTF-8 or not.
1740
1741                   •   Source filters activated within "eval" leak out into
1742                       whichever file scope is currently being compiled.  To
1743                       give an example with the CPAN module Semi::Semicolons:
1744
1745                        BEGIN { eval "use Semi::Semicolons; # not filtered" }
1746                        # filtered here!
1747
1748                       "evalbytes" fixes that to work the way one would
1749                       expect:
1750
1751                        use feature "evalbytes";
1752                        BEGIN { evalbytes "use Semi::Semicolons; # filtered" }
1753                        # not filtered
1754
1755               Problems can arise if the string expands a scalar containing a
1756               floating point number.  That scalar can expand to letters, such
1757               as "NaN" or "Infinity"; or, within the scope of a "use locale",
1758               the decimal point character may be something other than a dot
1759               (such as a comma).  None of these are likely to parse as you
1760               are likely expecting.
1761
1762               You should be especially careful to remember what's being
1763               looked at when:
1764
1765                   eval $x;        # CASE 1
1766                   eval "$x";      # CASE 2
1767
1768                   eval '$x';      # CASE 3
1769                   eval { $x };    # CASE 4
1770
1771                   eval "\$$x++";  # CASE 5
1772                   $$x++;          # CASE 6
1773
1774               Cases 1 and 2 above behave identically: they run the code
1775               contained in the variable $x.  (Although case 2 has misleading
1776               double quotes making the reader wonder what else might be
1777               happening (nothing is).)  Cases 3 and 4 likewise behave in the
1778               same way: they run the code '$x', which does nothing but return
1779               the value of $x.  (Case 4 is preferred for purely visual
1780               reasons, but it also has the advantage of compiling at compile-
1781               time instead of at run-time.)  Case 5 is a place where normally
1782               you would like to use double quotes, except that in this
1783               particular situation, you can just use symbolic references
1784               instead, as in case 6.
1785
1786               An "eval ''" executed within a subroutine defined in the "DB"
1787               package doesn't see the usual surrounding lexical scope, but
1788               rather the scope of the first non-DB piece of code that called
1789               it.  You don't normally need to worry about this unless you are
1790               writing a Perl debugger.
1791
1792               The final semicolon, if any, may be omitted from the value of
1793               EXPR.
1794
1795           Block eval
1796               If the code to be executed doesn't vary, you may use the eval-
1797               BLOCK form to trap run-time errors without incurring the
1798               penalty of recompiling each time.  The error, if any, is still
1799               returned in $@.  Examples:
1800
1801                   # make divide-by-zero nonfatal
1802                   eval { $answer = $a / $b; }; warn $@ if $@;
1803
1804                   # same thing, but less efficient
1805                   eval '$answer = $a / $b'; warn $@ if $@;
1806
1807                   # a compile-time error
1808                   eval { $answer = }; # WRONG
1809
1810                   # a run-time error
1811                   eval '$answer =';   # sets $@
1812
1813               If you want to trap errors when loading an XS module, some
1814               problems with the binary interface (such as Perl version skew)
1815               may be fatal even with "eval" unless $ENV{PERL_DL_NONLAZY} is
1816               set.  See perlrun.
1817
1818               Using the "eval {}" form as an exception trap in libraries does
1819               have some issues.  Due to the current arguably broken state of
1820               "__DIE__" hooks, you may wish not to trigger any "__DIE__"
1821               hooks that user code may have installed.  You can use the
1822               "local $SIG{__DIE__}" construct for this purpose, as this
1823               example shows:
1824
1825                   # a private exception trap for divide-by-zero
1826                   eval { local $SIG{'__DIE__'}; $answer = $a / $b; };
1827                   warn $@ if $@;
1828
1829               This is especially significant, given that "__DIE__" hooks can
1830               call "die" again, which has the effect of changing their error
1831               messages:
1832
1833                   # __DIE__ hooks may modify error messages
1834                   {
1835                      local $SIG{'__DIE__'} =
1836                             sub { (my $x = $_[0]) =~ s/foo/bar/g; die $x };
1837                      eval { die "foo lives here" };
1838                      print $@ if $@;                # prints "bar lives here"
1839                   }
1840
1841               Because this promotes action at a distance, this
1842               counterintuitive behavior may be fixed in a future release.
1843
1844               "eval BLOCK" does not count as a loop, so the loop control
1845               statements "next", "last", or "redo" cannot be used to leave or
1846               restart the block.
1847
1848               The final semicolon, if any, may be omitted from within the
1849               BLOCK.
1850
1851       evalbytes EXPR
1852       evalbytes
1853           This function is similar to a string eval, except it always parses
1854           its argument (or $_ if EXPR is omitted) as a string of independent
1855           bytes.
1856
1857           If called when "use utf8" is in effect, the string will be assumed
1858           to be encoded in UTF-8, and "evalbytes" will make a temporary copy
1859           to work from, downgraded to non-UTF-8.  If this is not possible
1860           (because one or more characters in it require UTF-8), the
1861           "evalbytes" will fail with the error stored in $@.
1862
1863           Bytes that correspond to ASCII-range code points will have their
1864           normal meanings for operators in the string.  The treatment of the
1865           other bytes depends on if the "'unicode_strings"" feature is in
1866           effect.
1867
1868           Of course, variables that are UTF-8 and are referred to in the
1869           string retain that:
1870
1871            my $a = "\x{100}";
1872            evalbytes 'print ord $a, "\n"';
1873
1874           prints
1875
1876            256
1877
1878           and $@ is empty.
1879
1880           Source filters activated within the evaluated code apply to the
1881           code itself.
1882
1883           "evalbytes" is available starting in Perl v5.16.  To access it, you
1884           must say "CORE::evalbytes", but you can omit the "CORE::" if the
1885           "evalbytes" feature is enabled.  This is enabled automatically with
1886           a "use v5.16" (or higher) declaration in the current scope.
1887
1888       exec LIST
1889       exec PROGRAM LIST
1890           The "exec" function executes a system command and never returns;
1891           use "system" instead of "exec" if you want it to return.  It fails
1892           and returns false only if the command does not exist and it is
1893           executed directly instead of via your system's command shell (see
1894           below).
1895
1896           Since it's a common mistake to use "exec" instead of "system", Perl
1897           warns you if "exec" is called in void context and if there is a
1898           following statement that isn't "die", "warn", or "exit" (if
1899           warnings are enabled--but you always do that, right?).  If you
1900           really want to follow an "exec" with some other statement, you can
1901           use one of these styles to avoid the warning:
1902
1903               exec ('foo')   or print STDERR "couldn't exec foo: $!";
1904               { exec ('foo') }; print STDERR "couldn't exec foo: $!";
1905
1906           If there is more than one argument in LIST, this calls execvp(3)
1907           with the arguments in LIST.  If there is only one element in LIST,
1908           the argument is checked for shell metacharacters, and if there are
1909           any, the entire argument is passed to the system's command shell
1910           for parsing (this is "/bin/sh -c" on Unix platforms, but varies on
1911           other platforms).  If there are no shell metacharacters in the
1912           argument, it is split into words and passed directly to "execvp",
1913           which is more efficient.  Examples:
1914
1915               exec '/bin/echo', 'Your arguments are: ', @ARGV;
1916               exec "sort $outfile | uniq";
1917
1918           If you don't really want to execute the first argument, but want to
1919           lie to the program you are executing about its own name, you can
1920           specify the program you actually want to run as an "indirect
1921           object" (without a comma) in front of the LIST, as in "exec PROGRAM
1922           LIST".  (This always forces interpretation of the LIST as a
1923           multivalued list, even if there is only a single scalar in the
1924           list.)  Example:
1925
1926               my $shell = '/bin/csh';
1927               exec $shell '-sh';    # pretend it's a login shell
1928
1929           or, more directly,
1930
1931               exec {'/bin/csh'} '-sh';  # pretend it's a login shell
1932
1933           When the arguments get executed via the system shell, results are
1934           subject to its quirks and capabilities.  See "`STRING`" in perlop
1935           for details.
1936
1937           Using an indirect object with "exec" or "system" is also more
1938           secure.  This usage (which also works fine with "system") forces
1939           interpretation of the arguments as a multivalued list, even if the
1940           list had just one argument.  That way you're safe from the shell
1941           expanding wildcards or splitting up words with whitespace in them.
1942
1943               my @args = ( "echo surprise" );
1944
1945               exec @args;               # subject to shell escapes
1946                                           # if @args == 1
1947               exec { $args[0] } @args;  # safe even with one-arg list
1948
1949           The first version, the one without the indirect object, ran the
1950           echo program, passing it "surprise" an argument.  The second
1951           version didn't; it tried to run a program named "echo surprise",
1952           didn't find it, and set $? to a non-zero value indicating failure.
1953
1954           On Windows, only the "exec PROGRAM LIST" indirect object syntax
1955           will reliably avoid using the shell; "exec LIST", even with more
1956           than one element, will fall back to the shell if the first spawn
1957           fails.
1958
1959           Perl attempts to flush all files opened for output before the exec,
1960           but this may not be supported on some platforms (see perlport).  To
1961           be safe, you may need to set $| ($AUTOFLUSH in English) or call the
1962           "autoflush" method of "IO::Handle" on any open handles to avoid
1963           lost output.
1964
1965           Note that "exec" will not call your "END" blocks, nor will it
1966           invoke "DESTROY" methods on your objects.
1967
1968           Portability issues: "exec" in perlport.
1969
1970       exists EXPR
1971           Given an expression that specifies an element of a hash, returns
1972           true if the specified element in the hash has ever been
1973           initialized, even if the corresponding value is undefined.
1974
1975               print "Exists\n"    if exists $hash{$key};
1976               print "Defined\n"   if defined $hash{$key};
1977               print "True\n"      if $hash{$key};
1978
1979           exists may also be called on array elements, but its behavior is
1980           much less obvious and is strongly tied to the use of "delete" on
1981           arrays.
1982
1983           WARNING: Calling "exists" on array values is strongly discouraged.
1984           The notion of deleting or checking the existence of Perl array
1985           elements is not conceptually coherent, and can lead to surprising
1986           behavior.
1987
1988               print "Exists\n"    if exists $array[$index];
1989               print "Defined\n"   if defined $array[$index];
1990               print "True\n"      if $array[$index];
1991
1992           A hash or array element can be true only if it's defined and
1993           defined only if it exists, but the reverse doesn't necessarily hold
1994           true.
1995
1996           Given an expression that specifies the name of a subroutine,
1997           returns true if the specified subroutine has ever been declared,
1998           even if it is undefined.  Mentioning a subroutine name for exists
1999           or defined does not count as declaring it.  Note that a subroutine
2000           that does not exist may still be callable: its package may have an
2001           "AUTOLOAD" method that makes it spring into existence the first
2002           time that it is called; see perlsub.
2003
2004               print "Exists\n"  if exists &subroutine;
2005               print "Defined\n" if defined &subroutine;
2006
2007           Note that the EXPR can be arbitrarily complicated as long as the
2008           final operation is a hash or array key lookup or subroutine name:
2009
2010               if (exists $ref->{A}->{B}->{$key})  { }
2011               if (exists $hash{A}{B}{$key})       { }
2012
2013               if (exists $ref->{A}->{B}->[$ix])   { }
2014               if (exists $hash{A}{B}[$ix])        { }
2015
2016               if (exists &{$ref->{A}{B}{$key}})   { }
2017
2018           Although the most deeply nested array or hash element will not
2019           spring into existence just because its existence was tested, any
2020           intervening ones will.  Thus "$ref->{"A"}" and "$ref->{"A"}->{"B"}"
2021           will spring into existence due to the existence test for the $key
2022           element above.  This happens anywhere the arrow operator is used,
2023           including even here:
2024
2025               undef $ref;
2026               if (exists $ref->{"Some key"})    { }
2027               print $ref;  # prints HASH(0x80d3d5c)
2028
2029           Use of a subroutine call, rather than a subroutine name, as an
2030           argument to "exists" is an error.
2031
2032               exists &sub;    # OK
2033               exists &sub();  # Error
2034
2035       exit EXPR
2036       exit
2037           Evaluates EXPR and exits immediately with that value.    Example:
2038
2039               my $ans = <STDIN>;
2040               exit 0 if $ans =~ /^[Xx]/;
2041
2042           See also "die".  If EXPR is omitted, exits with 0 status.  The only
2043           universally recognized values for EXPR are 0 for success and 1 for
2044           error; other values are subject to interpretation depending on the
2045           environment in which the Perl program is running.  For example,
2046           exiting 69 (EX_UNAVAILABLE) from a sendmail incoming-mail filter
2047           will cause the mailer to return the item undelivered, but that's
2048           not true everywhere.
2049
2050           Don't use "exit" to abort a subroutine if there's any chance that
2051           someone might want to trap whatever error happened.  Use "die"
2052           instead, which can be trapped by an "eval".
2053
2054           The "exit" function does not always exit immediately.  It calls any
2055           defined "END" routines first, but these "END" routines may not
2056           themselves abort the exit.  Likewise any object destructors that
2057           need to be called are called before the real exit.  "END" routines
2058           and destructors can change the exit status by modifying $?.  If
2059           this is a problem, you can call "POSIX::_exit($status)" to avoid
2060           "END" and destructor processing.  See perlmod for details.
2061
2062           Portability issues: "exit" in perlport.
2063
2064       exp EXPR
2065       exp Returns e (the natural logarithm base) to the power of EXPR.  If
2066           EXPR is omitted, gives "exp($_)".
2067
2068       fc EXPR
2069       fc  Returns the casefolded version of EXPR.  This is the internal
2070           function implementing the "\F" escape in double-quoted strings.
2071
2072           Casefolding is the process of mapping strings to a form where case
2073           differences are erased; comparing two strings in their casefolded
2074           form is effectively a way of asking if two strings are equal,
2075           regardless of case.
2076
2077           Roughly, if you ever found yourself writing this
2078
2079               lc($this) eq lc($that)    # Wrong!
2080                   # or
2081               uc($this) eq uc($that)    # Also wrong!
2082                   # or
2083               $this =~ /^\Q$that\E\z/i  # Right!
2084
2085           Now you can write
2086
2087               fc($this) eq fc($that)
2088
2089           And get the correct results.
2090
2091           Perl only implements the full form of casefolding, but you can
2092           access the simple folds using "casefold()" in Unicode::UCD and
2093           "prop_invmap()" in Unicode::UCD.  For further information on
2094           casefolding, refer to the Unicode Standard, specifically sections
2095           3.13 "Default Case Operations", 4.2 "Case-Normative", and 5.18
2096           "Case Mappings", available at
2097           <https://www.unicode.org/versions/latest/>, as well as the Case
2098           Charts available at <https://www.unicode.org/charts/case/>.
2099
2100           If EXPR is omitted, uses $_.
2101
2102           This function behaves the same way under various pragmas, such as
2103           within "use feature 'unicode_strings", as "lc" does, with the
2104           single exception of "fc" of LATIN CAPITAL LETTER SHARP S (U+1E9E)
2105           within the scope of "use locale".  The foldcase of this character
2106           would normally be "ss", but as explained in the "lc" section, case
2107           changes that cross the 255/256 boundary are problematic under
2108           locales, and are hence prohibited.  Therefore, this function under
2109           locale returns instead the string "\x{17F}\x{17F}", which is the
2110           LATIN SMALL LETTER LONG S.  Since that character itself folds to
2111           "s", the string of two of them together should be equivalent to a
2112           single U+1E9E when foldcased.
2113
2114           While the Unicode Standard defines two additional forms of
2115           casefolding, one for Turkic languages and one that never maps one
2116           character into multiple characters, these are not provided by the
2117           Perl core.  However, the CPAN module "Unicode::Casing" may be used
2118           to provide an implementation.
2119
2120           "fc" is available only if the "fc" feature is enabled or if it is
2121           prefixed with "CORE::".  The "fc" feature is enabled automatically
2122           with a "use v5.16" (or higher) declaration in the current scope.
2123
2124       fcntl FILEHANDLE,FUNCTION,SCALAR
2125           Implements the fcntl(2) function.  You'll probably have to say
2126
2127               use Fcntl;
2128
2129           first to get the correct constant definitions.  Argument processing
2130           and value returned work just like "ioctl" below.  For example:
2131
2132               use Fcntl;
2133               my $flags = fcntl($filehandle, F_GETFL, 0)
2134                   or die "Can't fcntl F_GETFL: $!";
2135
2136           You don't have to check for "defined" on the return from "fcntl".
2137           Like "ioctl", it maps a 0 return from the system call into "0 but
2138           true" in Perl.  This string is true in boolean context and 0 in
2139           numeric context.  It is also exempt from the normal "Argument "..."
2140           isn't numeric" warnings on improper numeric conversions.
2141
2142           Note that "fcntl" raises an exception if used on a machine that
2143           doesn't implement fcntl(2).  See the Fcntl module or your fcntl(2)
2144           manpage to learn what functions are available on your system.
2145
2146           Here's an example of setting a filehandle named $REMOTE to be non-
2147           blocking at the system level.  You'll have to negotiate $| on your
2148           own, though.
2149
2150               use Fcntl qw(F_GETFL F_SETFL O_NONBLOCK);
2151
2152               my $flags = fcntl($REMOTE, F_GETFL, 0)
2153                   or die "Can't get flags for the socket: $!\n";
2154
2155               fcntl($REMOTE, F_SETFL, $flags | O_NONBLOCK)
2156                   or die "Can't set flags for the socket: $!\n";
2157
2158           Portability issues: "fcntl" in perlport.
2159
2160       __FILE__
2161           A special token that returns the name of the file in which it
2162           occurs.  It can be altered by the mechanism described at "Plain Old
2163           Comments (Not!)" in perlsyn.
2164
2165       fileno FILEHANDLE
2166       fileno DIRHANDLE
2167           Returns the file descriptor for a filehandle or directory handle,
2168           or undefined if the filehandle is not open.  If there is no real
2169           file descriptor at the OS level, as can happen with filehandles
2170           connected to memory objects via "open" with a reference for the
2171           third argument, -1 is returned.
2172
2173           This is mainly useful for constructing bitmaps for "select" and
2174           low-level POSIX tty-handling operations.  If FILEHANDLE is an
2175           expression, the value is taken as an indirect filehandle, generally
2176           its name.
2177
2178           You can use this to find out whether two handles refer to the same
2179           underlying descriptor:
2180
2181               if (fileno($this) != -1 && fileno($this) == fileno($that)) {
2182                   print "\$this and \$that are dups\n";
2183               } elsif (fileno($this) != -1 && fileno($that) != -1) {
2184                   print "\$this and \$that have different " .
2185                       "underlying file descriptors\n";
2186               } else {
2187                   print "At least one of \$this and \$that does " .
2188                       "not have a real file descriptor\n";
2189               }
2190
2191           The behavior of "fileno" on a directory handle depends on the
2192           operating system.  On a system with dirfd(3) or similar, "fileno"
2193           on a directory handle returns the underlying file descriptor
2194           associated with the handle; on systems with no such support, it
2195           returns the undefined value, and sets $! (errno).
2196
2197       flock FILEHANDLE,OPERATION
2198           Calls flock(2), or an emulation of it, on FILEHANDLE.  Returns true
2199           for success, false on failure.  Produces a fatal error if used on a
2200           machine that doesn't implement flock(2), fcntl(2) locking, or
2201           lockf(3).  "flock" is Perl's portable file-locking interface,
2202           although it locks entire files only, not records.
2203
2204           Two potentially non-obvious but traditional "flock" semantics are
2205           that it waits indefinitely until the lock is granted, and that its
2206           locks are merely advisory.  Such discretionary locks are more
2207           flexible, but offer fewer guarantees.  This means that programs
2208           that do not also use "flock" may modify files locked with "flock".
2209           See perlport, your port's specific documentation, and your system-
2210           specific local manpages for details.  It's best to assume
2211           traditional behavior if you're writing portable programs.  (But if
2212           you're not, you should as always feel perfectly free to write for
2213           your own system's idiosyncrasies (sometimes called "features").
2214           Slavish adherence to portability concerns shouldn't get in the way
2215           of your getting your job done.)
2216
2217           OPERATION is one of LOCK_SH, LOCK_EX, or LOCK_UN, possibly combined
2218           with LOCK_NB.  These constants are traditionally valued 1, 2, 8 and
2219           4, but you can use the symbolic names if you import them from the
2220           Fcntl module, either individually, or as a group using the ":flock"
2221           tag.  LOCK_SH requests a shared lock, LOCK_EX requests an exclusive
2222           lock, and LOCK_UN releases a previously requested lock.  If LOCK_NB
2223           is bitwise-or'ed with LOCK_SH or LOCK_EX, then "flock" returns
2224           immediately rather than blocking waiting for the lock; check the
2225           return status to see if you got it.
2226
2227           To avoid the possibility of miscoordination, Perl now flushes
2228           FILEHANDLE before locking or unlocking it.
2229
2230           Note that the emulation built with lockf(3) doesn't provide shared
2231           locks, and it requires that FILEHANDLE be open with write intent.
2232           These are the semantics that lockf(3) implements.  Most if not all
2233           systems implement lockf(3) in terms of fcntl(2) locking, though, so
2234           the differing semantics shouldn't bite too many people.
2235
2236           Note that the fcntl(2) emulation of flock(3) requires that
2237           FILEHANDLE be open with read intent to use LOCK_SH and requires
2238           that it be open with write intent to use LOCK_EX.
2239
2240           Note also that some versions of "flock" cannot lock things over the
2241           network; you would need to use the more system-specific "fcntl" for
2242           that.  If you like you can force Perl to ignore your system's
2243           flock(2) function, and so provide its own fcntl(2)-based emulation,
2244           by passing the switch "-Ud_flock" to the Configure program when you
2245           configure and build a new Perl.
2246
2247           Here's a mailbox appender for BSD systems.
2248
2249               # import LOCK_* and SEEK_END constants
2250               use Fcntl qw(:flock SEEK_END);
2251
2252               sub lock {
2253                   my ($fh) = @_;
2254                   flock($fh, LOCK_EX) or die "Cannot lock mailbox - $!\n";
2255                   # and, in case we're running on a very old UNIX
2256                   # variant without the modern O_APPEND semantics...
2257                   seek($fh, 0, SEEK_END) or die "Cannot seek - $!\n";
2258               }
2259
2260               sub unlock {
2261                   my ($fh) = @_;
2262                   flock($fh, LOCK_UN) or die "Cannot unlock mailbox - $!\n";
2263               }
2264
2265               open(my $mbox, ">>", "/usr/spool/mail/$ENV{'USER'}")
2266                   or die "Can't open mailbox: $!";
2267
2268               lock($mbox);
2269               print $mbox $msg,"\n\n";
2270               unlock($mbox);
2271
2272           On systems that support a real flock(2), locks are inherited across
2273           "fork" calls, whereas those that must resort to the more capricious
2274           fcntl(2) function lose their locks, making it seriously harder to
2275           write servers.
2276
2277           See also DB_File for other "flock" examples.
2278
2279           Portability issues: "flock" in perlport.
2280
2281       fork
2282           Does a fork(2) system call to create a new process running the same
2283           program at the same point.  It returns the child pid to the parent
2284           process, 0 to the child process, or "undef" if the fork is
2285           unsuccessful.  File descriptors (and sometimes locks on those
2286           descriptors) are shared, while everything else is copied.  On most
2287           systems supporting fork(2), great care has gone into making it
2288           extremely efficient (for example, using copy-on-write technology on
2289           data pages), making it the dominant paradigm for multitasking over
2290           the last few decades.
2291
2292           Perl attempts to flush all files opened for output before forking
2293           the child process, but this may not be supported on some platforms
2294           (see perlport).  To be safe, you may need to set $| ($AUTOFLUSH in
2295           English) or call the "autoflush" method of "IO::Handle" on any open
2296           handles to avoid duplicate output.
2297
2298           If you "fork" without ever waiting on your children, you will
2299           accumulate zombies.  On some systems, you can avoid this by setting
2300           $SIG{CHLD} to "IGNORE".  See also perlipc for more examples of
2301           forking and reaping moribund children.
2302
2303           Note that if your forked child inherits system file descriptors
2304           like STDIN and STDOUT that are actually connected by a pipe or
2305           socket, even if you exit, then the remote server (such as, say, a
2306           CGI script or a backgrounded job launched from a remote shell)
2307           won't think you're done.  You should reopen those to /dev/null if
2308           it's any issue.
2309
2310           On some platforms such as Windows, where the fork(2) system call is
2311           not available, Perl can be built to emulate "fork" in the Perl
2312           interpreter.  The emulation is designed, at the level of the Perl
2313           program, to be as compatible as possible with the "Unix" fork(2).
2314           However it has limitations that have to be considered in code
2315           intended to be portable.  See perlfork for more details.
2316
2317           Portability issues: "fork" in perlport.
2318
2319       format
2320           Declare a picture format for use by the "write" function.  For
2321           example:
2322
2323               format Something =
2324                   Test: @<<<<<<<< @||||| @>>>>>
2325                         $str,     $%,    '$' . int($num)
2326               .
2327
2328               $str = "widget";
2329               $num = $cost/$quantity;
2330               $~ = 'Something';
2331               write;
2332
2333           See perlform for many details and examples.
2334
2335       formline PICTURE,LIST
2336           This is an internal function used by "format"s, though you may call
2337           it, too.  It formats (see perlform) a list of values according to
2338           the contents of PICTURE, placing the output into the format output
2339           accumulator, $^A (or $ACCUMULATOR in English).  Eventually, when a
2340           "write" is done, the contents of $^A are written to some
2341           filehandle.  You could also read $^A and then set $^A back to "".
2342           Note that a format typically does one "formline" per line of form,
2343           but the "formline" function itself doesn't care how many newlines
2344           are embedded in the PICTURE.  This means that the "~" and "~~"
2345           tokens treat the entire PICTURE as a single line.  You may
2346           therefore need to use multiple formlines to implement a single
2347           record format, just like the "format" compiler.
2348
2349           Be careful if you put double quotes around the picture, because an
2350           "@" character may be taken to mean the beginning of an array name.
2351           "formline" always returns true.  See perlform for other examples.
2352
2353           If you are trying to use this instead of "write" to capture the
2354           output, you may find it easier to open a filehandle to a scalar
2355           ("open my $fh, ">", \$output") and write to that instead.
2356
2357       getc FILEHANDLE
2358       getc
2359           Returns the next character from the input file attached to
2360           FILEHANDLE, or the undefined value at end of file or if there was
2361           an error (in the latter case $! is set).  If FILEHANDLE is omitted,
2362           reads from STDIN.  This is not particularly efficient.  However, it
2363           cannot be used by itself to fetch single characters without waiting
2364           for the user to hit enter.  For that, try something more like:
2365
2366               if ($BSD_STYLE) {
2367                   system "stty cbreak </dev/tty >/dev/tty 2>&1";
2368               }
2369               else {
2370                   system "stty", '-icanon', 'eol', "\001";
2371               }
2372
2373               my $key = getc(STDIN);
2374
2375               if ($BSD_STYLE) {
2376                   system "stty -cbreak </dev/tty >/dev/tty 2>&1";
2377               }
2378               else {
2379                   system 'stty', 'icanon', 'eol', '^@'; # ASCII NUL
2380               }
2381               print "\n";
2382
2383           Determination of whether $BSD_STYLE should be set is left as an
2384           exercise to the reader.
2385
2386           The "POSIX::getattr" function can do this more portably on systems
2387           purporting POSIX compliance.  See also the "Term::ReadKey" module
2388           on CPAN.
2389
2390       getlogin
2391           This implements the C library function of the same name, which on
2392           most systems returns the current login from /etc/utmp, if any.  If
2393           it returns the empty string, use "getpwuid".
2394
2395               my $login = getlogin || getpwuid($<) || "Kilroy";
2396
2397           Do not consider "getlogin" for authentication: it is not as secure
2398           as "getpwuid".
2399
2400           Portability issues: "getlogin" in perlport.
2401
2402       getpeername SOCKET
2403           Returns the packed sockaddr address of the other end of the SOCKET
2404           connection.
2405
2406               use Socket;
2407               my $hersockaddr    = getpeername($sock);
2408               my ($port, $iaddr) = sockaddr_in($hersockaddr);
2409               my $herhostname    = gethostbyaddr($iaddr, AF_INET);
2410               my $herstraddr     = inet_ntoa($iaddr);
2411
2412       getpgrp PID
2413           Returns the current process group for the specified PID.  Use a PID
2414           of 0 to get the current process group for the current process.
2415           Will raise an exception if used on a machine that doesn't implement
2416           getpgrp(2).  If PID is omitted, returns the process group of the
2417           current process.  Note that the POSIX version of "getpgrp" does not
2418           accept a PID argument, so only "PID==0" is truly portable.
2419
2420           Portability issues: "getpgrp" in perlport.
2421
2422       getppid
2423           Returns the process id of the parent process.
2424
2425           Note for Linux users: Between v5.8.1 and v5.16.0 Perl would work
2426           around non-POSIX thread semantics the minority of Linux systems
2427           (and Debian GNU/kFreeBSD systems) that used LinuxThreads, this
2428           emulation has since been removed.  See the documentation for $$ for
2429           details.
2430
2431           Portability issues: "getppid" in perlport.
2432
2433       getpriority WHICH,WHO
2434           Returns the current priority for a process, a process group, or a
2435           user.  (See getpriority(2).)  Will raise a fatal exception if used
2436           on a machine that doesn't implement getpriority(2).
2437
2438           "WHICH" can be any of "PRIO_PROCESS", "PRIO_PGRP" or "PRIO_USER"
2439           imported from "RESOURCE CONSTANTS" in POSIX.
2440
2441           Portability issues: "getpriority" in perlport.
2442
2443       getpwnam NAME
2444       getgrnam NAME
2445       gethostbyname NAME
2446       getnetbyname NAME
2447       getprotobyname NAME
2448       getpwuid UID
2449       getgrgid GID
2450       getservbyname NAME,PROTO
2451       gethostbyaddr ADDR,ADDRTYPE
2452       getnetbyaddr ADDR,ADDRTYPE
2453       getprotobynumber NUMBER
2454       getservbyport PORT,PROTO
2455       getpwent
2456       getgrent
2457       gethostent
2458       getnetent
2459       getprotoent
2460       getservent
2461       setpwent
2462       setgrent
2463       sethostent STAYOPEN
2464       setnetent STAYOPEN
2465       setprotoent STAYOPEN
2466       setservent STAYOPEN
2467       endpwent
2468       endgrent
2469       endhostent
2470       endnetent
2471       endprotoent
2472       endservent
2473           These routines are the same as their counterparts in the system C
2474           library.  In list context, the return values from the various get
2475           routines are as follows:
2476
2477            #    0        1          2           3         4
2478            my ( $name,   $passwd,   $gid,       $members  ) = getgr*
2479            my ( $name,   $aliases,  $addrtype,  $net      ) = getnet*
2480            my ( $name,   $aliases,  $port,      $proto    ) = getserv*
2481            my ( $name,   $aliases,  $proto                ) = getproto*
2482            my ( $name,   $aliases,  $addrtype,  $length,  @addrs ) = gethost*
2483            my ( $name,   $passwd,   $uid,       $gid,     $quota,
2484               $comment,  $gcos,     $dir,       $shell,   $expire ) = getpw*
2485            #    5        6          7           8         9
2486
2487           (If the entry doesn't exist, the return value is a single
2488           meaningless true value.)
2489
2490           The exact meaning of the $gcos field varies but it usually contains
2491           the real name of the user (as opposed to the login name) and other
2492           information pertaining to the user.  Beware, however, that in many
2493           system users are able to change this information and therefore it
2494           cannot be trusted and therefore the $gcos is tainted (see perlsec).
2495           The $passwd and $shell, user's encrypted password and login shell,
2496           are also tainted, for the same reason.
2497
2498           In scalar context, you get the name, unless the function was a
2499           lookup by name, in which case you get the other thing, whatever it
2500           is.  (If the entry doesn't exist you get the undefined value.)  For
2501           example:
2502
2503               my $uid   = getpwnam($name);
2504               my $name  = getpwuid($num);
2505               my $name  = getpwent();
2506               my $gid   = getgrnam($name);
2507               my $name  = getgrgid($num);
2508               my $name  = getgrent();
2509               # etc.
2510
2511           In getpw*() the fields $quota, $comment, and $expire are special in
2512           that they are unsupported on many systems.  If the $quota is
2513           unsupported, it is an empty scalar.  If it is supported, it usually
2514           encodes the disk quota.  If the $comment field is unsupported, it
2515           is an empty scalar.  If it is supported it usually encodes some
2516           administrative comment about the user.  In some systems the $quota
2517           field may be $change or $age, fields that have to do with password
2518           aging.  In some systems the $comment field may be $class.  The
2519           $expire field, if present, encodes the expiration period of the
2520           account or the password.  For the availability and the exact
2521           meaning of these fields in your system, please consult getpwnam(3)
2522           and your system's pwd.h file.  You can also find out from within
2523           Perl what your $quota and $comment fields mean and whether you have
2524           the $expire field by using the "Config" module and the values
2525           "d_pwquota", "d_pwage", "d_pwchange", "d_pwcomment", and
2526           "d_pwexpire".  Shadow password files are supported only if your
2527           vendor has implemented them in the intuitive fashion that calling
2528           the regular C library routines gets the shadow versions if you're
2529           running under privilege or if there exists the shadow(3) functions
2530           as found in System V (this includes Solaris and Linux).  Those
2531           systems that implement a proprietary shadow password facility are
2532           unlikely to be supported.
2533
2534           The $members value returned by getgr*() is a space-separated list
2535           of the login names of the members of the group.
2536
2537           For the gethost*() functions, if the "h_errno" variable is
2538           supported in C, it will be returned to you via $? if the function
2539           call fails.  The @addrs value returned by a successful call is a
2540           list of raw addresses returned by the corresponding library call.
2541           In the Internet domain, each address is four bytes long; you can
2542           unpack it by saying something like:
2543
2544               my ($w,$x,$y,$z) = unpack('W4',$addr[0]);
2545
2546           The Socket library makes this slightly easier:
2547
2548               use Socket;
2549               my $iaddr = inet_aton("127.1"); # or whatever address
2550               my $name  = gethostbyaddr($iaddr, AF_INET);
2551
2552               # or going the other way
2553               my $straddr = inet_ntoa($iaddr);
2554
2555           In the opposite way, to resolve a hostname to the IP address you
2556           can write this:
2557
2558               use Socket;
2559               my $packed_ip = gethostbyname("www.perl.org");
2560               my $ip_address;
2561               if (defined $packed_ip) {
2562                   $ip_address = inet_ntoa($packed_ip);
2563               }
2564
2565           Make sure "gethostbyname" is called in SCALAR context and that its
2566           return value is checked for definedness.
2567
2568           The "getprotobynumber" function, even though it only takes one
2569           argument, has the precedence of a list operator, so beware:
2570
2571               getprotobynumber $number eq 'icmp'   # WRONG
2572               getprotobynumber($number eq 'icmp')  # actually means this
2573               getprotobynumber($number) eq 'icmp'  # better this way
2574
2575           If you get tired of remembering which element of the return list
2576           contains which return value, by-name interfaces are provided in
2577           standard modules: "File::stat", "Net::hostent", "Net::netent",
2578           "Net::protoent", "Net::servent", "Time::gmtime", "Time::localtime",
2579           and "User::grent".  These override the normal built-ins, supplying
2580           versions that return objects with the appropriate names for each
2581           field.  For example:
2582
2583              use File::stat;
2584              use User::pwent;
2585              my $is_his = (stat($filename)->uid == pwent($whoever)->uid);
2586
2587           Even though it looks as though they're the same method calls (uid),
2588           they aren't, because a "File::stat" object is different from a
2589           "User::pwent" object.
2590
2591           Many of these functions are not safe in a multi-threaded
2592           environment where more than one thread can be using them.  In
2593           particular, functions like "getpwent()" iterate per-process and not
2594           per-thread, so if two threads are simultaneously iterating, neither
2595           will get all the records.
2596
2597           Some systems have thread-safe versions of some of the functions,
2598           such as "getpwnam_r()" instead of "getpwnam()".  There, Perl
2599           automatically and invisibly substitutes the thread-safe version,
2600           without notice.  This means that code that safely runs on some
2601           systems can fail on others that lack the thread-safe versions.
2602
2603           Portability issues: "getpwnam" in perlport to "endservent" in
2604           perlport.
2605
2606       getsockname SOCKET
2607           Returns the packed sockaddr address of this end of the SOCKET
2608           connection, in case you don't know the address because you have
2609           several different IPs that the connection might have come in on.
2610
2611               use Socket;
2612               my $mysockaddr = getsockname($sock);
2613               my ($port, $myaddr) = sockaddr_in($mysockaddr);
2614               printf "Connect to %s [%s]\n",
2615                  scalar gethostbyaddr($myaddr, AF_INET),
2616                  inet_ntoa($myaddr);
2617
2618       getsockopt SOCKET,LEVEL,OPTNAME
2619           Queries the option named OPTNAME associated with SOCKET at a given
2620           LEVEL.  Options may exist at multiple protocol levels depending on
2621           the socket type, but at least the uppermost socket level SOL_SOCKET
2622           (defined in the "Socket" module) will exist.  To query options at
2623           another level the protocol number of the appropriate protocol
2624           controlling the option should be supplied.  For example, to
2625           indicate that an option is to be interpreted by the TCP protocol,
2626           LEVEL should be set to the protocol number of TCP, which you can
2627           get using "getprotobyname".
2628
2629           The function returns a packed string representing the requested
2630           socket option, or "undef" on error, with the reason for the error
2631           placed in $!.  Just what is in the packed string depends on LEVEL
2632           and OPTNAME; consult getsockopt(2) for details.  A common case is
2633           that the option is an integer, in which case the result is a packed
2634           integer, which you can decode using "unpack" with the "i" (or "I")
2635           format.
2636
2637           Here's an example to test whether Nagle's algorithm is enabled on a
2638           socket:
2639
2640               use Socket qw(:all);
2641
2642               defined(my $tcp = getprotobyname("tcp"))
2643                   or die "Could not determine the protocol number for tcp";
2644               # my $tcp = IPPROTO_TCP; # Alternative
2645               my $packed = getsockopt($socket, $tcp, TCP_NODELAY)
2646                   or die "getsockopt TCP_NODELAY: $!";
2647               my $nodelay = unpack("I", $packed);
2648               print "Nagle's algorithm is turned ",
2649                      $nodelay ? "off\n" : "on\n";
2650
2651           Portability issues: "getsockopt" in perlport.
2652
2653       glob EXPR
2654       glob
2655           In list context, returns a (possibly empty) list of filename
2656           expansions on the value of EXPR such as the standard Unix shell
2657           /bin/csh would do.  In scalar context, glob iterates through such
2658           filename expansions, returning undef when the list is exhausted.
2659           This is the internal function implementing the "<*.c>" operator,
2660           but you can use it directly.  If EXPR is omitted, $_ is used.  The
2661           "<*.c>" operator is discussed in more detail in "I/O Operators" in
2662           perlop.
2663
2664           Note that "glob" splits its arguments on whitespace and treats each
2665           segment as separate pattern.  As such, "glob("*.c *.h")" matches
2666           all files with a .c or .h extension.  The expression "glob(".* *")"
2667           matches all files in the current working directory.  If you want to
2668           glob filenames that might contain whitespace, you'll have to use
2669           extra quotes around the spacey filename to protect it.  For
2670           example, to glob filenames that have an "e" followed by a space
2671           followed by an "f", use one of:
2672
2673               my @spacies = <"*e f*">;
2674               my @spacies = glob '"*e f*"';
2675               my @spacies = glob q("*e f*");
2676
2677           If you had to get a variable through, you could do this:
2678
2679               my @spacies = glob "'*${var}e f*'";
2680               my @spacies = glob qq("*${var}e f*");
2681
2682           If non-empty braces are the only wildcard characters used in the
2683           "glob", no filenames are matched, but potentially many strings are
2684           returned.  For example, this produces nine strings, one for each
2685           pairing of fruits and colors:
2686
2687               my @many = glob "{apple,tomato,cherry}={green,yellow,red}";
2688
2689           This operator is implemented using the standard "File::Glob"
2690           extension.  See File::Glob for details, including "bsd_glob", which
2691           does not treat whitespace as a pattern separator.
2692
2693           If a "glob" expression is used as the condition of a "while" or
2694           "for" loop, then it will be implicitly assigned to $_.  If either a
2695           "glob" expression or an explicit assignment of a "glob" expression
2696           to a scalar is used as a "while"/"for" condition, then the
2697           condition actually tests for definedness of the expression's value,
2698           not for its regular truth value.
2699
2700           Portability issues: "glob" in perlport.
2701
2702       gmtime EXPR
2703       gmtime
2704           Works just like "localtime" but the returned values are localized
2705           for the standard Greenwich time zone.
2706
2707           Note: When called in list context, $isdst, the last value returned
2708           by gmtime, is always 0.  There is no Daylight Saving Time in GMT.
2709
2710           Portability issues: "gmtime" in perlport.
2711
2712       goto LABEL
2713       goto EXPR
2714       goto &NAME
2715           The "goto LABEL" form finds the statement labeled with LABEL and
2716           resumes execution there.  It can't be used to get out of a block or
2717           subroutine given to "sort".  It can be used to go almost anywhere
2718           else within the dynamic scope, including out of subroutines, but
2719           it's usually better to use some other construct such as "last" or
2720           "die".  The author of Perl has never felt the need to use this form
2721           of "goto" (in Perl, that is; C is another matter).  (The difference
2722           is that C does not offer named loops combined with loop control.
2723           Perl does, and this replaces most structured uses of "goto" in
2724           other languages.)
2725
2726           The "goto EXPR" form expects to evaluate "EXPR" to a code reference
2727           or a label name.  If it evaluates to a code reference, it will be
2728           handled like "goto &NAME", below.  This is especially useful for
2729           implementing tail recursion via "goto __SUB__".
2730
2731           If the expression evaluates to a label name, its scope will be
2732           resolved dynamically.  This allows for computed "goto"s per
2733           FORTRAN, but isn't necessarily recommended if you're optimizing for
2734           maintainability:
2735
2736               goto ("FOO", "BAR", "GLARCH")[$i];
2737
2738           As shown in this example, "goto EXPR" is exempt from the "looks
2739           like a function" rule.  A pair of parentheses following it does not
2740           (necessarily) delimit its argument.  "goto("NE")."XT"" is
2741           equivalent to "goto NEXT".  Also, unlike most named operators, this
2742           has the same precedence as assignment.
2743
2744           Use of "goto LABEL" or "goto EXPR" to jump into a construct is
2745           deprecated and will issue a warning.  Even then, it may not be used
2746           to go into any construct that requires initialization, such as a
2747           subroutine, a "foreach" loop, or a "given" block.  In general, it
2748           may not be used to jump into the parameter of a binary or list
2749           operator, but it may be used to jump into the first parameter of a
2750           binary operator.  (The "=" assignment operator's "first" operand is
2751           its right-hand operand.)  It also can't be used to go into a
2752           construct that is optimized away.
2753
2754           The "goto &NAME" form is quite different from the other forms of
2755           "goto".  In fact, it isn't a goto in the normal sense at all, and
2756           doesn't have the stigma associated with other gotos.  Instead, it
2757           exits the current subroutine (losing any changes set by "local")
2758           and immediately calls in its place the named subroutine using the
2759           current value of @_.  This is used by "AUTOLOAD" subroutines that
2760           wish to load another subroutine and then pretend that the other
2761           subroutine had been called in the first place (except that any
2762           modifications to @_ in the current subroutine are propagated to the
2763           other subroutine.) After the "goto", not even "caller" will be able
2764           to tell that this routine was called first.
2765
2766           NAME needn't be the name of a subroutine; it can be a scalar
2767           variable containing a code reference or a block that evaluates to a
2768           code reference.
2769
2770       grep BLOCK LIST
2771       grep EXPR,LIST
2772           This is similar in spirit to, but not the same as, grep(1) and its
2773           relatives.  In particular, it is not limited to using regular
2774           expressions.
2775
2776           Evaluates the BLOCK or EXPR for each element of LIST (locally
2777           setting $_ to each element) and returns the list value consisting
2778           of those elements for which the expression evaluated to true.  In
2779           scalar context, returns the number of times the expression was
2780           true.
2781
2782               my @foo = grep(!/^#/, @bar);    # weed out comments
2783
2784           or equivalently,
2785
2786               my @foo = grep {!/^#/} @bar;    # weed out comments
2787
2788           Note that $_ is an alias to the list value, so it can be used to
2789           modify the elements of the LIST.  While this is useful and
2790           supported, it can cause bizarre results if the elements of LIST are
2791           not variables.  Similarly, grep returns aliases into the original
2792           list, much as a for loop's index variable aliases the list
2793           elements.  That is, modifying an element of a list returned by grep
2794           (for example, in a "foreach", "map" or another "grep") actually
2795           modifies the element in the original list.  This is usually
2796           something to be avoided when writing clear code.
2797
2798           See also "map" for a list composed of the results of the BLOCK or
2799           EXPR.
2800
2801       hex EXPR
2802       hex Interprets EXPR as a hex string and returns the corresponding
2803           numeric value.  If EXPR is omitted, uses $_.
2804
2805               print hex '0xAf'; # prints '175'
2806               print hex 'aF';   # same
2807               $valid_input =~ /\A(?:0?[xX])?(?:_?[0-9a-fA-F])*\z/
2808
2809           A hex string consists of hex digits and an optional "0x" or "x"
2810           prefix.  Each hex digit may be preceded by a single underscore,
2811           which will be ignored.  Any other character triggers a warning and
2812           causes the rest of the string to be ignored (even leading
2813           whitespace, unlike "oct").  Only integers can be represented, and
2814           integer overflow triggers a warning.
2815
2816           To convert strings that might start with any of 0, "0x", or "0b",
2817           see "oct".  To present something as hex, look into "printf",
2818           "sprintf", and "unpack".
2819
2820       import LIST
2821           There is no builtin "import" function.  It is just an ordinary
2822           method (subroutine) defined (or inherited) by modules that wish to
2823           export names to another module.  The "use" function calls the
2824           "import" method for the package used.  See also "use", perlmod, and
2825           Exporter.
2826
2827       index STR,SUBSTR,POSITION
2828       index STR,SUBSTR
2829           The index function searches for one string within another, but
2830           without the wildcard-like behavior of a full regular-expression
2831           pattern match.  It returns the position of the first occurrence of
2832           SUBSTR in STR at or after POSITION.  If POSITION is omitted, starts
2833           searching from the beginning of the string.  POSITION before the
2834           beginning of the string or after its end is treated as if it were
2835           the beginning or the end, respectively.  POSITION and the return
2836           value are based at zero.  If the substring is not found, "index"
2837           returns -1.
2838
2839       int EXPR
2840       int Returns the integer portion of EXPR.  If EXPR is omitted, uses $_.
2841           You should not use this function for rounding: one because it
2842           truncates towards 0, and two because machine representations of
2843           floating-point numbers can sometimes produce counterintuitive
2844           results.  For example, "int(-6.725/0.025)" produces -268 rather
2845           than the correct -269; that's because it's really more like
2846           -268.99999999999994315658 instead.  Usually, the "sprintf",
2847           "printf", or the "POSIX::floor" and "POSIX::ceil" functions will
2848           serve you better than will "int".
2849
2850       ioctl FILEHANDLE,FUNCTION,SCALAR
2851           Implements the ioctl(2) function.  You'll probably first have to
2852           say
2853
2854               require "sys/ioctl.ph";  # probably in
2855                                        # $Config{archlib}/sys/ioctl.ph
2856
2857           to get the correct function definitions.  If sys/ioctl.ph doesn't
2858           exist or doesn't have the correct definitions you'll have to roll
2859           your own, based on your C header files such as <sys/ioctl.h>.
2860           (There is a Perl script called h2ph that comes with the Perl kit
2861           that may help you in this, but it's nontrivial.)  SCALAR will be
2862           read and/or written depending on the FUNCTION; a C pointer to the
2863           string value of SCALAR will be passed as the third argument of the
2864           actual "ioctl" call.  (If SCALAR has no string value but does have
2865           a numeric value, that value will be passed rather than a pointer to
2866           the string value.  To guarantee this to be true, add a 0 to the
2867           scalar before using it.)  The "pack" and "unpack" functions may be
2868           needed to manipulate the values of structures used by "ioctl".
2869
2870           The return value of "ioctl" (and "fcntl") is as follows:
2871
2872               if OS returns:      then Perl returns:
2873                   -1               undefined value
2874                    0              string "0 but true"
2875               anything else           that number
2876
2877           Thus Perl returns true on success and false on failure, yet you can
2878           still easily determine the actual value returned by the operating
2879           system:
2880
2881               my $retval = ioctl(...) || -1;
2882               printf "System returned %d\n", $retval;
2883
2884           The special string "0 but true" is exempt from "Argument "..."
2885           isn't numeric" warnings on improper numeric conversions.
2886
2887           Portability issues: "ioctl" in perlport.
2888
2889       join EXPR,LIST
2890           Joins the separate strings of LIST into a single string with fields
2891           separated by the value of EXPR, and returns that new string.
2892           Example:
2893
2894              my $rec = join(':', $login,$passwd,$uid,$gid,$gcos,$home,$shell);
2895
2896           Beware that unlike "split", "join" doesn't take a pattern as its
2897           first argument.  Compare "split".
2898
2899       keys HASH
2900       keys ARRAY
2901           Called in list context, returns a list consisting of all the keys
2902           of the named hash, or in Perl 5.12 or later only, the indices of an
2903           array.  Perl releases prior to 5.12 will produce a syntax error if
2904           you try to use an array argument.  In scalar context, returns the
2905           number of keys or indices.
2906
2907           Hash entries are returned in an apparently random order.  The
2908           actual random order is specific to a given hash; the exact same
2909           series of operations on two hashes may result in a different order
2910           for each hash.  Any insertion into the hash may change the order,
2911           as will any deletion, with the exception that the most recent key
2912           returned by "each" or "keys" may be deleted without changing the
2913           order.  So long as a given hash is unmodified you may rely on
2914           "keys", "values" and "each" to repeatedly return the same order as
2915           each other.  See "Algorithmic Complexity Attacks" in perlsec for
2916           details on why hash order is randomized.  Aside from the guarantees
2917           provided here the exact details of Perl's hash algorithm and the
2918           hash traversal order are subject to change in any release of Perl.
2919           Tied hashes may behave differently to Perl's hashes with respect to
2920           changes in order on insertion and deletion of items.
2921
2922           As a side effect, calling "keys" resets the internal iterator of
2923           the HASH or ARRAY (see "each") before yielding the keys.  In
2924           particular, calling "keys" in void context resets the iterator with
2925           no other overhead.
2926
2927           Here is yet another way to print your environment:
2928
2929               my @keys = keys %ENV;
2930               my @values = values %ENV;
2931               while (@keys) {
2932                   print pop(@keys), '=', pop(@values), "\n";
2933               }
2934
2935           or how about sorted by key:
2936
2937               foreach my $key (sort(keys %ENV)) {
2938                   print $key, '=', $ENV{$key}, "\n";
2939               }
2940
2941           The returned values are copies of the original keys in the hash, so
2942           modifying them will not affect the original hash.  Compare
2943           "values".
2944
2945           To sort a hash by value, you'll need to use a "sort" function.
2946           Here's a descending numeric sort of a hash by its values:
2947
2948               foreach my $key (sort { $hash{$b} <=> $hash{$a} } keys %hash) {
2949                   printf "%4d %s\n", $hash{$key}, $key;
2950               }
2951
2952           Used as an lvalue, "keys" allows you to increase the number of hash
2953           buckets allocated for the given hash.  This can gain you a measure
2954           of efficiency if you know the hash is going to get big.  (This is
2955           similar to pre-extending an array by assigning a larger number to
2956           $#array.)  If you say
2957
2958               keys %hash = 200;
2959
2960           then %hash will have at least 200 buckets allocated for it--256 of
2961           them, in fact, since it rounds up to the next power of two.  These
2962           buckets will be retained even if you do "%hash = ()", use "undef
2963           %hash" if you want to free the storage while %hash is still in
2964           scope.  You can't shrink the number of buckets allocated for the
2965           hash using "keys" in this way (but you needn't worry about doing
2966           this by accident, as trying has no effect).  "keys @array" in an
2967           lvalue context is a syntax error.
2968
2969           Starting with Perl 5.14, an experimental feature allowed "keys" to
2970           take a scalar expression. This experiment has been deemed
2971           unsuccessful, and was removed as of Perl 5.24.
2972
2973           To avoid confusing would-be users of your code who are running
2974           earlier versions of Perl with mysterious syntax errors, put this
2975           sort of thing at the top of your file to signal that your code will
2976           work only on Perls of a recent vintage:
2977
2978               use 5.012;  # so keys/values/each work on arrays
2979
2980           See also "each", "values", and "sort".
2981
2982       kill SIGNAL, LIST
2983       kill SIGNAL
2984           Sends a signal to a list of processes.  Returns the number of
2985           arguments that were successfully used to signal (which is not
2986           necessarily the same as the number of processes actually killed,
2987           e.g. where a process group is killed).
2988
2989               my $cnt = kill 'HUP', $child1, $child2;
2990               kill 'KILL', @goners;
2991
2992           SIGNAL may be either a signal name (a string) or a signal number.
2993           A signal name may start with a "SIG" prefix, thus "FOO" and
2994           "SIGFOO" refer to the same signal.  The string form of SIGNAL is
2995           recommended for portability because the same signal may have
2996           different numbers in different operating systems.
2997
2998           A list of signal names supported by the current platform can be
2999           found in $Config{sig_name}, which is provided by the "Config"
3000           module.  See Config for more details.
3001
3002           A negative signal name is the same as a negative signal number,
3003           killing process groups instead of processes.  For example, "kill
3004           '-KILL', $pgrp" and "kill -9, $pgrp" will send "SIGKILL" to the
3005           entire process group specified.  That means you usually want to use
3006           positive not negative signals.
3007
3008           If SIGNAL is either the number 0 or the string "ZERO" (or
3009           "SIGZERO"), no signal is sent to the process, but "kill" checks
3010           whether it's possible to send a signal to it (that means, to be
3011           brief, that the process is owned by the same user, or we are the
3012           super-user).  This is useful to check that a child process is still
3013           alive (even if only as a zombie) and hasn't changed its UID.  See
3014           perlport for notes on the portability of this construct.
3015
3016           The behavior of kill when a PROCESS number is zero or negative
3017           depends on the operating system.  For example, on POSIX-conforming
3018           systems, zero will signal the current process group, -1 will signal
3019           all processes, and any other negative PROCESS number will act as a
3020           negative signal number and kill the entire process group specified.
3021
3022           If both the SIGNAL and the PROCESS are negative, the results are
3023           undefined.  A warning may be produced in a future version.
3024
3025           See "Signals" in perlipc for more details.
3026
3027           On some platforms such as Windows where the fork(2) system call is
3028           not available, Perl can be built to emulate "fork" at the
3029           interpreter level.  This emulation has limitations related to kill
3030           that have to be considered, for code running on Windows and in code
3031           intended to be portable.
3032
3033           See perlfork for more details.
3034
3035           If there is no LIST of processes, no signal is sent, and the return
3036           value is 0.  This form is sometimes used, however, because it
3037           causes tainting checks to be run.  But see "Laundering and
3038           Detecting Tainted Data" in perlsec.
3039
3040           Portability issues: "kill" in perlport.
3041
3042       last LABEL
3043       last EXPR
3044       last
3045           The "last" command is like the "break" statement in C (as used in
3046           loops); it immediately exits the loop in question.  If the LABEL is
3047           omitted, the command refers to the innermost enclosing loop.  The
3048           "last EXPR" form, available starting in Perl 5.18.0, allows a label
3049           name to be computed at run time, and is otherwise identical to
3050           "last LABEL".  The "continue" block, if any, is not executed:
3051
3052               LINE: while (<STDIN>) {
3053                   last LINE if /^$/;  # exit when done with header
3054                   #...
3055               }
3056
3057           "last" cannot return a value from a block that typically returns a
3058           value, such as "eval {}", "sub {}", or "do {}". It will perform its
3059           flow control behavior, which precludes any return value. It should
3060           not be used to exit a "grep" or "map" operation.
3061
3062           Note that a block by itself is semantically identical to a loop
3063           that executes once.  Thus "last" can be used to effect an early
3064           exit out of such a block.
3065
3066           See also "continue" for an illustration of how "last", "next", and
3067           "redo" work.
3068
3069           Unlike most named operators, this has the same precedence as
3070           assignment.  It is also exempt from the looks-like-a-function rule,
3071           so "last ("foo")."bar"" will cause "bar" to be part of the argument
3072           to "last".
3073
3074       lc EXPR
3075       lc  Returns a lowercased version of EXPR.  This is the internal
3076           function implementing the "\L" escape in double-quoted strings.
3077
3078           If EXPR is omitted, uses $_.
3079
3080           What gets returned depends on several factors:
3081
3082           If "use bytes" is in effect:
3083               The results follow ASCII rules.  Only the characters "A-Z"
3084               change, to "a-z" respectively.
3085
3086           Otherwise, if "use locale" for "LC_CTYPE" is in effect:
3087               Respects current "LC_CTYPE" locale for code points < 256; and
3088               uses Unicode rules for the remaining code points (this last can
3089               only happen if the UTF8 flag is also set).  See perllocale.
3090
3091               Starting in v5.20, Perl uses full Unicode rules if the locale
3092               is UTF-8.  Otherwise, there is a deficiency in this scheme,
3093               which is that case changes that cross the 255/256 boundary are
3094               not well-defined.  For example, the lower case of LATIN CAPITAL
3095               LETTER SHARP S (U+1E9E) in Unicode rules is U+00DF (on ASCII
3096               platforms).   But under "use locale" (prior to v5.20 or not a
3097               UTF-8 locale), the lower case of U+1E9E is itself, because 0xDF
3098               may not be LATIN SMALL LETTER SHARP S in the current locale,
3099               and Perl has no way of knowing if that character even exists in
3100               the locale, much less what code point it is.  Perl returns a
3101               result that is above 255 (almost always the input character
3102               unchanged), for all instances (and there aren't many) where the
3103               255/256 boundary would otherwise be crossed; and starting in
3104               v5.22, it raises a locale warning.
3105
3106           Otherwise, If EXPR has the UTF8 flag set:
3107               Unicode rules are used for the case change.
3108
3109           Otherwise, if "use feature 'unicode_strings'" or "use locale
3110           ':not_characters'" is in effect:
3111               Unicode rules are used for the case change.
3112
3113           Otherwise:
3114               ASCII rules are used for the case change.  The lowercase of any
3115               character outside the ASCII range is the character itself.
3116
3117       lcfirst EXPR
3118       lcfirst
3119           Returns the value of EXPR with the first character lowercased.
3120           This is the internal function implementing the "\l" escape in
3121           double-quoted strings.
3122
3123           If EXPR is omitted, uses $_.
3124
3125           This function behaves the same way under various pragmas, such as
3126           in a locale, as "lc" does.
3127
3128       length EXPR
3129       length
3130           Returns the length in characters of the value of EXPR.  If EXPR is
3131           omitted, returns the length of $_.  If EXPR is undefined, returns
3132           "undef".
3133
3134           This function cannot be used on an entire array or hash to find out
3135           how many elements these have.  For that, use "scalar @array" and
3136           "scalar keys %hash", respectively.
3137
3138           Like all Perl character operations, "length" normally deals in
3139           logical characters, not physical bytes.  For how many bytes a
3140           string encoded as UTF-8 would take up, use
3141           "length(Encode::encode('UTF-8', EXPR))" (you'll have to "use
3142           Encode" first).  See Encode and perlunicode.
3143
3144       __LINE__
3145           A special token that compiles to the current line number.  It can
3146           be altered by the mechanism described at "Plain Old Comments
3147           (Not!)" in perlsyn.
3148
3149       link OLDFILE,NEWFILE
3150           Creates a new filename linked to the old filename.  Returns true
3151           for success, false otherwise.
3152
3153           Portability issues: "link" in perlport.
3154
3155       listen SOCKET,QUEUESIZE
3156           Does the same thing that the listen(2) system call does.  Returns
3157           true if it succeeded, false otherwise.  See the example in
3158           "Sockets: Client/Server Communication" in perlipc.
3159
3160       local EXPR
3161           You really probably want to be using "my" instead, because "local"
3162           isn't what most people think of as "local".  See "Private Variables
3163           via my()" in perlsub for details.
3164
3165           A local modifies the listed variables to be local to the enclosing
3166           block, file, or eval.  If more than one value is listed, the list
3167           must be placed in parentheses.  See "Temporary Values via local()"
3168           in perlsub for details, including issues with tied arrays and
3169           hashes.
3170
3171           The "delete local EXPR" construct can also be used to localize the
3172           deletion of array/hash elements to the current block.  See
3173           "Localized deletion of elements of composite types" in perlsub.
3174
3175       localtime EXPR
3176       localtime
3177           Converts a time as returned by the time function to a 9-element
3178           list with the time analyzed for the local time zone.  Typically
3179           used as follows:
3180
3181               #     0    1    2     3     4    5     6     7     8
3182               my ($sec,$min,$hour,$mday,$mon,$year,$wday,$yday,$isdst) =
3183                                                           localtime(time);
3184
3185           All list elements are numeric and come straight out of the C
3186           `struct tm'.  $sec, $min, and $hour are the seconds, minutes, and
3187           hours of the specified time.
3188
3189           $mday is the day of the month and $mon the month in the range
3190           0..11, with 0 indicating January and 11 indicating December.  This
3191           makes it easy to get a month name from a list:
3192
3193               my @abbr = qw(Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec);
3194               print "$abbr[$mon] $mday";
3195               # $mon=9, $mday=18 gives "Oct 18"
3196
3197           $year contains the number of years since 1900.  To get a 4-digit
3198           year write:
3199
3200               $year += 1900;
3201
3202           To get the last two digits of the year (e.g., "01" in 2001) do:
3203
3204               $year = sprintf("%02d", $year % 100);
3205
3206           $wday is the day of the week, with 0 indicating Sunday and 3
3207           indicating Wednesday.  $yday is the day of the year, in the range
3208           0..364 (or 0..365 in leap years.)
3209
3210           $isdst is true if the specified time occurs during Daylight Saving
3211           Time, false otherwise.
3212
3213           If EXPR is omitted, "localtime" uses the current time (as returned
3214           by "time").
3215
3216           In scalar context, "localtime" returns the ctime(3) value:
3217
3218               my $now_string = localtime;  # e.g., "Thu Oct 13 04:54:34 1994"
3219
3220           The format of this scalar value is not locale-dependent but built
3221           into Perl.  For GMT instead of local time use the "gmtime" builtin.
3222           See also the "Time::Local" module (for converting seconds, minutes,
3223           hours, and such back to the integer value returned by "time"), and
3224           the POSIX module's "strftime" and "mktime" functions.
3225
3226           To get somewhat similar but locale-dependent date strings, set up
3227           your locale environment variables appropriately (please see
3228           perllocale) and try for example:
3229
3230               use POSIX qw(strftime);
3231               my $now_string = strftime "%a %b %e %H:%M:%S %Y", localtime;
3232               # or for GMT formatted appropriately for your locale:
3233               my $now_string = strftime "%a %b %e %H:%M:%S %Y", gmtime;
3234
3235           Note that %a and %b, the short forms of the day of the week and the
3236           month of the year, may not necessarily be three characters wide.
3237
3238           The Time::gmtime and Time::localtime modules provide a convenient,
3239           by-name access mechanism to the "gmtime" and "localtime" functions,
3240           respectively.
3241
3242           For a comprehensive date and time representation look at the
3243           DateTime module on CPAN.
3244
3245           Portability issues: "localtime" in perlport.
3246
3247       lock THING
3248           This function places an advisory lock on a shared variable or
3249           referenced object contained in THING until the lock goes out of
3250           scope.
3251
3252           The value returned is the scalar itself, if the argument is a
3253           scalar, or a reference, if the argument is a hash, array or
3254           subroutine.
3255
3256           "lock" is a "weak keyword"; this means that if you've defined a
3257           function by this name (before any calls to it), that function will
3258           be called instead.  If you are not under "use threads::shared" this
3259           does nothing.  See threads::shared.
3260
3261       log EXPR
3262       log Returns the natural logarithm (base e) of EXPR.  If EXPR is
3263           omitted, returns the log of $_.  To get the log of another base,
3264           use basic algebra: The base-N log of a number is equal to the
3265           natural log of that number divided by the natural log of N.  For
3266           example:
3267
3268               sub log10 {
3269                   my $n = shift;
3270                   return log($n)/log(10);
3271               }
3272
3273           See also "exp" for the inverse operation.
3274
3275       lstat FILEHANDLE
3276       lstat EXPR
3277       lstat DIRHANDLE
3278       lstat
3279           Does the same thing as the "stat" function (including setting the
3280           special "_" filehandle) but stats a symbolic link instead of the
3281           file the symbolic link points to.  If symbolic links are
3282           unimplemented on your system, a normal "stat" is done.  For much
3283           more detailed information, please see the documentation for "stat".
3284
3285           If EXPR is omitted, stats $_.
3286
3287           Portability issues: "lstat" in perlport.
3288
3289       m// The match operator.  See "Regexp Quote-Like Operators" in perlop.
3290
3291       map BLOCK LIST
3292       map EXPR,LIST
3293           Evaluates the BLOCK or EXPR for each element of LIST (locally
3294           setting $_ to each element) and composes a list of the results of
3295           each such evaluation.  Each element of LIST may produce zero, one,
3296           or more elements in the generated list, so the number of elements
3297           in the generated list may differ from that in LIST.  In scalar
3298           context, returns the total number of elements so generated.  In
3299           list context, returns the generated list.
3300
3301               my @chars = map(chr, @numbers);
3302
3303           translates a list of numbers to the corresponding characters.
3304
3305               my @squares = map { $_ * $_ } @numbers;
3306
3307           translates a list of numbers to their squared values.
3308
3309               my @squares = map { $_ > 5 ? ($_ * $_) : () } @numbers;
3310
3311           shows that number of returned elements can differ from the number
3312           of input elements.  To omit an element, return an empty list ().
3313           This could also be achieved by writing
3314
3315               my @squares = map { $_ * $_ } grep { $_ > 5 } @numbers;
3316
3317           which makes the intention more clear.
3318
3319           Map always returns a list, which can be assigned to a hash such
3320           that the elements become key/value pairs.  See perldata for more
3321           details.
3322
3323               my %hash = map { get_a_key_for($_) => $_ } @array;
3324
3325           is just a funny way to write
3326
3327               my %hash;
3328               foreach (@array) {
3329                   $hash{get_a_key_for($_)} = $_;
3330               }
3331
3332           Note that $_ is an alias to the list value, so it can be used to
3333           modify the elements of the LIST.  While this is useful and
3334           supported, it can cause bizarre results if the elements of LIST are
3335           not variables.  Using a regular "foreach" loop for this purpose
3336           would be clearer in most cases.  See also "grep" for a list
3337           composed of those items of the original list for which the BLOCK or
3338           EXPR evaluates to true.
3339
3340           "{" starts both hash references and blocks, so "map { ..." could be
3341           either the start of map BLOCK LIST or map EXPR, LIST.  Because Perl
3342           doesn't look ahead for the closing "}" it has to take a guess at
3343           which it's dealing with based on what it finds just after the "{".
3344           Usually it gets it right, but if it doesn't it won't realize
3345           something is wrong until it gets to the "}" and encounters the
3346           missing (or unexpected) comma.  The syntax error will be reported
3347           close to the "}", but you'll need to change something near the "{"
3348           such as using a unary "+" or semicolon to give Perl some help:
3349
3350            my %hash = map {  "\L$_" => 1  } @array # perl guesses EXPR. wrong
3351            my %hash = map { +"\L$_" => 1  } @array # perl guesses BLOCK. right
3352            my %hash = map {; "\L$_" => 1  } @array # this also works
3353            my %hash = map { ("\L$_" => 1) } @array # as does this
3354            my %hash = map {  lc($_) => 1  } @array # and this.
3355            my %hash = map +( lc($_) => 1 ), @array # this is EXPR and works!
3356
3357            my %hash = map  ( lc($_), 1 ),   @array # evaluates to (1, @array)
3358
3359           or to force an anon hash constructor use "+{":
3360
3361               my @hashes = map +{ lc($_) => 1 }, @array # EXPR, so needs
3362                                                         # comma at end
3363
3364           to get a list of anonymous hashes each with only one entry apiece.
3365
3366       mkdir FILENAME,MODE
3367       mkdir FILENAME
3368       mkdir
3369           Creates the directory specified by FILENAME, with permissions
3370           specified by MODE (as modified by "umask").  If it succeeds it
3371           returns true; otherwise it returns false and sets $! (errno).  MODE
3372           defaults to 0777 if omitted, and FILENAME defaults to $_ if
3373           omitted.
3374
3375           In general, it is better to create directories with a permissive
3376           MODE and let the user modify that with their "umask" than it is to
3377           supply a restrictive MODE and give the user no way to be more
3378           permissive.  The exceptions to this rule are when the file or
3379           directory should be kept private (mail files, for instance).  The
3380           documentation for "umask" discusses the choice of MODE in more
3381           detail.
3382
3383           Note that according to the POSIX 1003.1-1996 the FILENAME may have
3384           any number of trailing slashes.  Some operating and filesystems do
3385           not get this right, so Perl automatically removes all trailing
3386           slashes to keep everyone happy.
3387
3388           To recursively create a directory structure, look at the
3389           "make_path" function of the File::Path module.
3390
3391       msgctl ID,CMD,ARG
3392           Calls the System V IPC function msgctl(2).  You'll probably have to
3393           say
3394
3395               use IPC::SysV;
3396
3397           first to get the correct constant definitions.  If CMD is
3398           "IPC_STAT", then ARG must be a variable that will hold the returned
3399           "msqid_ds" structure.  Returns like "ioctl": the undefined value
3400           for error, "0 but true" for zero, or the actual return value
3401           otherwise.  See also "SysV IPC" in perlipc and the documentation
3402           for "IPC::SysV" and "IPC::Semaphore".
3403
3404           Portability issues: "msgctl" in perlport.
3405
3406       msgget KEY,FLAGS
3407           Calls the System V IPC function msgget(2).  Returns the message
3408           queue id, or "undef" on error.  See also "SysV IPC" in perlipc and
3409           the documentation for "IPC::SysV" and "IPC::Msg".
3410
3411           Portability issues: "msgget" in perlport.
3412
3413       msgrcv ID,VAR,SIZE,TYPE,FLAGS
3414           Calls the System V IPC function msgrcv to receive a message from
3415           message queue ID into variable VAR with a maximum message size of
3416           SIZE.  Note that when a message is received, the message type as a
3417           native long integer will be the first thing in VAR, followed by the
3418           actual message.  This packing may be opened with "unpack("l! a*")".
3419           Taints the variable.  Returns true if successful, false on error.
3420           See also "SysV IPC" in perlipc and the documentation for
3421           "IPC::SysV" and "IPC::Msg".
3422
3423           Portability issues: "msgrcv" in perlport.
3424
3425       msgsnd ID,MSG,FLAGS
3426           Calls the System V IPC function msgsnd to send the message MSG to
3427           the message queue ID.  MSG must begin with the native long integer
3428           message type, be followed by the length of the actual message, and
3429           then finally the message itself.  This kind of packing can be
3430           achieved with "pack("l! a*", $type, $message)".  Returns true if
3431           successful, false on error.  See also "SysV IPC" in perlipc and the
3432           documentation for "IPC::SysV" and "IPC::Msg".
3433
3434           Portability issues: "msgsnd" in perlport.
3435
3436       my VARLIST
3437       my TYPE VARLIST
3438       my VARLIST : ATTRS
3439       my TYPE VARLIST : ATTRS
3440           A "my" declares the listed variables to be local (lexically) to the
3441           enclosing block, file, or "eval".  If more than one variable is
3442           listed, the list must be placed in parentheses.
3443
3444           The exact semantics and interface of TYPE and ATTRS are still
3445           evolving.  TYPE may be a bareword, a constant declared with "use
3446           constant", or "__PACKAGE__".  It is currently bound to the use of
3447           the fields pragma, and attributes are handled using the attributes
3448           pragma, or starting from Perl 5.8.0 also via the
3449           Attribute::Handlers module.  See "Private Variables via my()" in
3450           perlsub for details.
3451
3452           Note that with a parenthesised list, "undef" can be used as a dummy
3453           placeholder, for example to skip assignment of initial values:
3454
3455               my ( undef, $min, $hour ) = localtime;
3456
3457       next LABEL
3458       next EXPR
3459       next
3460           The "next" command is like the "continue" statement in C; it starts
3461           the next iteration of the loop:
3462
3463               LINE: while (<STDIN>) {
3464                   next LINE if /^#/;  # discard comments
3465                   #...
3466               }
3467
3468           Note that if there were a "continue" block on the above, it would
3469           get executed even on discarded lines.  If LABEL is omitted, the
3470           command refers to the innermost enclosing loop.  The "next EXPR"
3471           form, available as of Perl 5.18.0, allows a label name to be
3472           computed at run time, being otherwise identical to "next LABEL".
3473
3474           "next" cannot return a value from a block that typically returns a
3475           value, such as "eval {}", "sub {}", or "do {}". It will perform its
3476           flow control behavior, which precludes any return value. It should
3477           not be used to exit a "grep" or "map" operation.
3478
3479           Note that a block by itself is semantically identical to a loop
3480           that executes once.  Thus "next" will exit such a block early.
3481
3482           See also "continue" for an illustration of how "last", "next", and
3483           "redo" work.
3484
3485           Unlike most named operators, this has the same precedence as
3486           assignment.  It is also exempt from the looks-like-a-function rule,
3487           so "next ("foo")."bar"" will cause "bar" to be part of the argument
3488           to "next".
3489
3490       no MODULE VERSION LIST
3491       no MODULE VERSION
3492       no MODULE LIST
3493       no MODULE
3494       no VERSION
3495           See the "use" function, of which "no" is the opposite.
3496
3497       oct EXPR
3498       oct Interprets EXPR as an octal string and returns the corresponding
3499           value.  (If EXPR happens to start off with "0x", interprets it as a
3500           hex string.  If EXPR starts off with "0b", it is interpreted as a
3501           binary string.  Leading whitespace is ignored in all three cases.)
3502           The following will handle decimal, binary, octal, and hex in
3503           standard Perl notation:
3504
3505               $val = oct($val) if $val =~ /^0/;
3506
3507           If EXPR is omitted, uses $_.   To go the other way (produce a
3508           number in octal), use "sprintf" or "printf":
3509
3510               my $dec_perms = (stat("filename"))[2] & 07777;
3511               my $oct_perm_str = sprintf "%o", $perms;
3512
3513           The "oct" function is commonly used when a string such as 644 needs
3514           to be converted into a file mode, for example.  Although Perl
3515           automatically converts strings into numbers as needed, this
3516           automatic conversion assumes base 10.
3517
3518           Leading white space is ignored without warning, as too are any
3519           trailing non-digits, such as a decimal point ("oct" only handles
3520           non-negative integers, not negative integers or floating point).
3521
3522       open FILEHANDLE,MODE,EXPR
3523       open FILEHANDLE,MODE,EXPR,LIST
3524       open FILEHANDLE,MODE,REFERENCE
3525       open FILEHANDLE,EXPR
3526       open FILEHANDLE
3527           Associates an internal FILEHANDLE with the external file specified
3528           by EXPR. That filehandle will subsequently allow you to perform I/O
3529           operations on that file, such as reading from it or writing to it.
3530
3531           Instead of a filename, you may specify an external command (plus an
3532           optional argument list) or a scalar reference, in order to open
3533           filehandles on commands or in-memory scalars, respectively.
3534
3535           A thorough reference to "open" follows. For a gentler introduction
3536           to the basics of "open", see also the perlopentut manual page.
3537
3538           Working with files
3539               Most often, "open" gets invoked with three arguments: the
3540               required FILEHANDLE (usually an empty scalar variable),
3541               followed by MODE (usually a literal describing the I/O mode the
3542               filehandle will use), and then the filename  that the new
3543               filehandle will refer to.
3544
3545               Simple examples
3546                   Reading from a file:
3547
3548                       open(my $fh, "<", "input.txt")
3549                           or die "Can't open < input.txt: $!";
3550
3551                       # Process every line in input.txt
3552                       while (my $line = <$fh>) {
3553                           #
3554                           # ... do something interesting with $line here ...
3555                           #
3556                       }
3557
3558                   or writing to one:
3559
3560                       open(my $fh, ">", "output.txt")
3561                           or die "Can't open > output.txt: $!";
3562
3563                       print $fh "This line gets printed into output.txt.\n";
3564
3565                   For a summary of common filehandle operations such as
3566                   these, see "Files and I/O" in perlintro.
3567
3568               About filehandles
3569                   The first argument to "open", labeled FILEHANDLE in this
3570                   reference, is usually a scalar variable. (Exceptions exist,
3571                   described in "Other considerations", below.) If the call to
3572                   "open" succeeds, then the expression provided as FILEHANDLE
3573                   will get assigned an open filehandle. That filehandle
3574                   provides an internal reference to the specified external
3575                   file, conveniently stored in a Perl variable, and ready for
3576                   I/O operations such as reading and writing.
3577
3578               About modes
3579                   When calling "open" with three or more arguments, the
3580                   second argument -- labeled MODE here -- defines the open
3581                   mode. MODE is usually a literal string comprising special
3582                   characters that define the intended I/O role of the
3583                   filehandle being created: whether it's read-only, or read-
3584                   and-write, and so on.
3585
3586                   If MODE is "<", the file is opened for input (read-only).
3587                   If MODE is ">", the file is opened for output, with
3588                   existing files first being truncated ("clobbered") and
3589                   nonexisting files newly created.  If MODE is ">>", the file
3590                   is opened for appending, again being created if necessary.
3591
3592                   You can put a "+" in front of the ">" or "<" to indicate
3593                   that you want both read and write access to the file; thus
3594                   "+<" is almost always preferred for read/write updates--the
3595                   "+>" mode would clobber the file first.  You can't usually
3596                   use either read-write mode for updating textfiles, since
3597                   they have variable-length records.  See the -i switch in
3598                   perlrun for a better approach.  The file is created with
3599                   permissions of 0666 modified by the process's "umask"
3600                   value.
3601
3602                   These various prefixes correspond to the fopen(3) modes of
3603                   "r", "r+", "w", "w+", "a", and "a+".
3604
3605                   More examples of different modes in action:
3606
3607                    # Open a file for concatenation
3608                    open(my $log, ">>", "/usr/spool/news/twitlog")
3609                        or warn "Couldn't open log file; discarding input";
3610
3611                    # Open a file for reading and writing
3612                    open(my $dbase, "+<", "dbase.mine")
3613                        or die "Can't open 'dbase.mine' for update: $!";
3614
3615               Checking the return value
3616                   Open returns nonzero on success, the undefined value
3617                   otherwise.  If the "open" involved a pipe, the return value
3618                   happens to be the pid of the subprocess.
3619
3620                   When opening a file, it's seldom a good idea to continue if
3621                   the request failed, so "open" is frequently used with
3622                   "die". Even if you want your code to do something other
3623                   than "die" on a failed open, you should still always check
3624                   the return value from opening a file.
3625
3626           Specifying I/O layers in MODE
3627               You can use the three-argument form of open to specify I/O
3628               layers (sometimes referred to as "disciplines") to apply to the
3629               new filehandle. These affect how the input and output are
3630               processed (see open and PerlIO for more details).  For example:
3631
3632                   open(my $fh, "<:encoding(UTF-8)", $filename)
3633                       || die "Can't open UTF-8 encoded $filename: $!";
3634
3635               This opens the UTF8-encoded file containing Unicode characters;
3636               see perluniintro.  Note that if layers are specified in the
3637               three-argument form, then default layers stored in "${^OPEN}"
3638               (usually set by the open pragma or the switch "-CioD") are
3639               ignored.  Those layers will also be ignored if you specify a
3640               colon with no name following it.  In that case the default
3641               layer for the operating system (:raw on Unix, :crlf on Windows)
3642               is used.
3643
3644               On some systems (in general, DOS- and Windows-based systems)
3645               "binmode" is necessary when you're not working with a text
3646               file.  For the sake of portability it is a good idea always to
3647               use it when appropriate, and never to use it when it isn't
3648               appropriate.  Also, people can set their I/O to be by default
3649               UTF8-encoded Unicode, not bytes.
3650
3651           Using "undef" for temporary files
3652               As a special case the three-argument form with a read/write
3653               mode and the third argument being "undef":
3654
3655                   open(my $tmp, "+>", undef) or die ...
3656
3657               opens a filehandle to a newly created empty anonymous temporary
3658               file.  (This happens under any mode, which makes "+>" the only
3659               useful and sensible mode to use.)  You will need to "seek" to
3660               do the reading.
3661
3662           Opening a filehandle into an in-memory scalar
3663               You can open filehandles directly to Perl scalars instead of a
3664               file or other resource external to the program. To do so,
3665               provide a reference to that scalar as the third argument to
3666               "open", like so:
3667
3668                open(my $memory, ">", \$var)
3669                    or die "Can't open memory file: $!";
3670                print $memory "foo!\n";    # output will appear in $var
3671
3672               To (re)open "STDOUT" or "STDERR" as an in-memory file, close it
3673               first:
3674
3675                   close STDOUT;
3676                   open(STDOUT, ">", \$variable)
3677                       or die "Can't open STDOUT: $!";
3678
3679               The scalars for in-memory files are treated as octet strings:
3680               unless the file is being opened with truncation the scalar may
3681               not contain any code points over 0xFF.
3682
3683               Opening in-memory files can fail for a variety of reasons.  As
3684               with any other "open", check the return value for success.
3685
3686               Technical note: This feature works only when Perl is built with
3687               PerlIO -- the default, except with older (pre-5.16) Perl
3688               installations that were configured to not include it (e.g. via
3689               "Configure -Uuseperlio"). You can see whether your Perl was
3690               built with PerlIO by running "perl -V:useperlio".  If it says
3691               'define', you have PerlIO; otherwise you don't.
3692
3693               See perliol for detailed info on PerlIO.
3694
3695           Opening a filehandle into a command
3696               If MODE is "|-", then the filename is interpreted as a command
3697               to which output is to be piped, and if MODE is "-|", the
3698               filename is interpreted as a command that pipes output to us.
3699               In the two-argument (and one-argument) form, one should replace
3700               dash ("-") with the command.  See "Using open() for IPC" in
3701               perlipc for more examples of this.  (You are not allowed to
3702               "open" to a command that pipes both in and out, but see
3703               IPC::Open2, IPC::Open3, and "Bidirectional Communication with
3704               Another Process" in perlipc for alternatives.)
3705
3706                open(my $article_fh, "-|", "caesar <$article")  # decrypt
3707                                                                # article
3708                    or die "Can't start caesar: $!";
3709
3710                open(my $article_fh, "caesar <$article |")      # ditto
3711                    or die "Can't start caesar: $!";
3712
3713                open(my $out_fh, "|-", "sort >Tmp$$")    # $$ is our process id
3714                    or die "Can't start sort: $!";
3715
3716               In the form of pipe opens taking three or more arguments, if
3717               LIST is specified (extra arguments after the command name) then
3718               LIST becomes arguments to the command invoked if the platform
3719               supports it.  The meaning of "open" with more than three
3720               arguments for non-pipe modes is not yet defined, but
3721               experimental "layers" may give extra LIST arguments meaning.
3722
3723               If you open a pipe on the command "-" (that is, specify either
3724               "|-" or "-|" with the one- or two-argument forms of "open"), an
3725               implicit "fork" is done, so "open" returns twice: in the parent
3726               process it returns the pid of the child process, and in the
3727               child process it returns (a defined) 0.  Use "defined($pid)" or
3728               "//" to determine whether the open was successful.
3729
3730               For example, use either
3731
3732                  my $child_pid = open(my $from_kid, "-|")
3733                       // die "Can't fork: $!";
3734
3735               or
3736
3737                  my $child_pid = open(my $to_kid,   "|-")
3738                       // die "Can't fork: $!";
3739
3740               followed by
3741
3742                   if ($child_pid) {
3743                       # am the parent:
3744                       # either write $to_kid or else read $from_kid
3745                       ...
3746                      waitpid $child_pid, 0;
3747                   } else {
3748                       # am the child; use STDIN/STDOUT normally
3749                       ...
3750                       exit;
3751                   }
3752
3753               The filehandle behaves normally for the parent, but I/O to that
3754               filehandle is piped from/to the STDOUT/STDIN of the child
3755               process.  In the child process, the filehandle isn't
3756               opened--I/O happens from/to the new STDOUT/STDIN.  Typically
3757               this is used like the normal piped open when you want to
3758               exercise more control over just how the pipe command gets
3759               executed, such as when running setuid and you don't want to
3760               have to scan shell commands for metacharacters.
3761
3762               The following blocks are more or less equivalent:
3763
3764                   open(my $fh, "|tr '[a-z]' '[A-Z]'");
3765                   open(my $fh, "|-", "tr '[a-z]' '[A-Z]'");
3766                   open(my $fh, "|-") || exec 'tr', '[a-z]', '[A-Z]';
3767                   open(my $fh, "|-", "tr", '[a-z]', '[A-Z]');
3768
3769                   open(my $fh, "cat -n '$file'|");
3770                   open(my $fh, "-|", "cat -n '$file'");
3771                   open(my $fh, "-|") || exec "cat", "-n", $file;
3772                   open(my $fh, "-|", "cat", "-n", $file);
3773
3774               The last two examples in each block show the pipe as "list
3775               form", which is not yet supported on all platforms. (If your
3776               platform has a real "fork", such as Linux and macOS, you can
3777               use the list form; it also works on Windows with Perl 5.22 or
3778               later.) You would want to use the list form of the pipe so you
3779               can pass literal arguments to the command without risk of the
3780               shell interpreting any shell metacharacters in them. However,
3781               this also bars you from opening pipes to commands that
3782               intentionally contain shell metacharacters, such as:
3783
3784                   open(my $fh, "|cat -n | expand -4 | lpr")
3785                       || die "Can't open pipeline to lpr: $!";
3786
3787               See "Safe Pipe Opens" in perlipc for more examples of this.
3788
3789           Duping filehandles
3790               You may also, in the Bourne shell tradition, specify an EXPR
3791               beginning with ">&", in which case the rest of the string is
3792               interpreted as the name of a filehandle (or file descriptor, if
3793               numeric) to be duped (as in dup(2)) and opened.  You may use
3794               "&" after ">", ">>", "<", "+>", "+>>", and "+<".  The mode you
3795               specify should match the mode of the original filehandle.
3796               (Duping a filehandle does not take into account any existing
3797               contents of IO buffers.)  If you use the three-argument form,
3798               then you can pass either a number, the name of a filehandle, or
3799               the normal "reference to a glob".
3800
3801               Here is a script that saves, redirects, and restores "STDOUT"
3802               and "STDERR" using various methods:
3803
3804                   #!/usr/bin/perl
3805                   open(my $oldout, ">&STDOUT")
3806                       or die "Can't dup STDOUT: $!";
3807                   open(OLDERR,     ">&", \*STDERR)
3808                       or die "Can't dup STDERR: $!";
3809
3810                   open(STDOUT, '>', "foo.out")
3811                       or die "Can't redirect STDOUT: $!";
3812                   open(STDERR, ">&STDOUT")
3813                       or die "Can't dup STDOUT: $!";
3814
3815                   select STDERR; $| = 1;  # make unbuffered
3816                   select STDOUT; $| = 1;  # make unbuffered
3817
3818                   print STDOUT "stdout 1\n";  # this works for
3819                   print STDERR "stderr 1\n";  # subprocesses too
3820
3821                   open(STDOUT, ">&", $oldout)
3822                       or die "Can't dup \$oldout: $!";
3823                   open(STDERR, ">&OLDERR")
3824                       or die "Can't dup OLDERR: $!";
3825
3826                   print STDOUT "stdout 2\n";
3827                   print STDERR "stderr 2\n";
3828
3829               If you specify '<&=X', where "X" is a file descriptor number or
3830               a filehandle, then Perl will do an equivalent of C's fdopen(3)
3831               of that file descriptor (and not call dup(2)); this is more
3832               parsimonious of file descriptors.  For example:
3833
3834                   # open for input, reusing the fileno of $fd
3835                   open(my $fh, "<&=", $fd)
3836
3837               or
3838
3839                   open(my $fh, "<&=$fd")
3840
3841               or
3842
3843                   # open for append, using the fileno of $oldfh
3844                   open(my $fh, ">>&=", $oldfh)
3845
3846               Being parsimonious on filehandles is also useful (besides being
3847               parsimonious) for example when something is dependent on file
3848               descriptors, like for example locking using "flock".  If you do
3849               just "open(my $A, ">>&", $B)", the filehandle $A will not have
3850               the same file descriptor as $B, and therefore "flock($A)" will
3851               not "flock($B)" nor vice versa.  But with "open(my $A, ">>&=",
3852               $B)", the filehandles will share the same underlying system
3853               file descriptor.
3854
3855               Note that under Perls older than 5.8.0, Perl uses the standard
3856               C library's' fdopen(3) to implement the "=" functionality.  On
3857               many Unix systems, fdopen(3) fails when file descriptors exceed
3858               a certain value, typically 255.  For Perls 5.8.0 and later,
3859               PerlIO is (most often) the default.
3860
3861           Legacy usage
3862               This section describes ways to call "open" outside of best
3863               practices; you may encounter these uses in older code. Perl
3864               does not consider their use deprecated, exactly, but neither is
3865               it recommended in new code, for the sake of clarity and
3866               readability.
3867
3868               Specifying mode and filename as a single argument
3869                   In the one- and two-argument forms of the call, the mode
3870                   and filename should be concatenated (in that order),
3871                   preferably separated by white space.  You can--but
3872                   shouldn't--omit the mode in these forms when that mode is
3873                   "<".  It is safe to use the two-argument form of "open" if
3874                   the filename argument is a known literal.
3875
3876                    open(my $dbase, "+<dbase.mine")          # ditto
3877                        or die "Can't open 'dbase.mine' for update: $!";
3878
3879                   In the two-argument (and one-argument) form, opening "<-"
3880                   or "-" opens STDIN and opening ">-" opens STDOUT.
3881
3882                   New code should favor the three-argument form of "open"
3883                   over this older form. Declaring the mode and the filename
3884                   as two distinct arguments avoids any confusion between the
3885                   two.
3886
3887               Calling "open" with one argument via global variables
3888                   As a shortcut, a one-argument call takes the filename from
3889                   the global scalar variable of the same name as the
3890                   filehandle:
3891
3892                       $ARTICLE = 100;
3893                       open(ARTICLE)
3894                           or die "Can't find article $ARTICLE: $!\n";
3895
3896                   Here $ARTICLE must be a global (package) scalar variable -
3897                   not one declared with "my" or "state".
3898
3899               Assigning a filehandle to a bareword
3900                   An older style is to use a bareword as the filehandle, as
3901
3902                       open(FH, "<", "input.txt")
3903                          or die "Can't open < input.txt: $!";
3904
3905                   Then you can use "FH" as the filehandle, in "close FH" and
3906                   "<FH>" and so on.  Note that it's a global variable, so
3907                   this form is not recommended when dealing with filehandles
3908                   other than Perl's built-in ones (e.g. STDOUT and STDIN).
3909
3910           Other considerations
3911               Automatic filehandle closure
3912                   The filehandle will be closed when its reference count
3913                   reaches zero. If it is a lexically scoped variable declared
3914                   with "my", that usually means the end of the enclosing
3915                   scope.  However, this automatic close does not check for
3916                   errors, so it is better to explicitly close filehandles,
3917                   especially those used for writing:
3918
3919                       close($handle)
3920                          || warn "close failed: $!";
3921
3922               Automatic pipe flushing
3923                   Perl will attempt to flush all files opened for output
3924                   before any operation that may do a fork, but this may not
3925                   be supported on some platforms (see perlport).  To be safe,
3926                   you may need to set $| ($AUTOFLUSH in English) or call the
3927                   "autoflush" method of "IO::Handle" on any open handles.
3928
3929                   On systems that support a close-on-exec flag on files, the
3930                   flag will be set for the newly opened file descriptor as
3931                   determined by the value of $^F.  See "$^F" in perlvar.
3932
3933                   Closing any piped filehandle causes the parent process to
3934                   wait for the child to finish, then returns the status value
3935                   in $? and "${^CHILD_ERROR_NATIVE}".
3936
3937               Direct versus by-reference assignment of filehandles
3938                   If FILEHANDLE -- the first argument in a call to "open" --
3939                   is an undefined scalar variable (or array or hash element),
3940                   a new filehandle is autovivified, meaning that the variable
3941                   is assigned a reference to a newly allocated anonymous
3942                   filehandle.  Otherwise if FILEHANDLE is an expression, its
3943                   value is the real filehandle.  (This is considered a
3944                   symbolic reference, so "use strict "refs"" should not be in
3945                   effect.)
3946
3947               Whitespace and special characters in the filename argument
3948                   The filename passed to the one- and two-argument forms of
3949                   "open" will have leading and trailing whitespace deleted
3950                   and normal redirection characters honored.  This property,
3951                   known as "magic open", can often be used to good effect.  A
3952                   user could specify a filename of "rsh cat file |", or you
3953                   could change certain filenames as needed:
3954
3955                       $filename =~ s/(.*\.gz)\s*$/gzip -dc < $1|/;
3956                       open(my $fh, $filename)
3957                           or die "Can't open $filename: $!";
3958
3959                   Use the three-argument form to open a file with arbitrary
3960                   weird characters in it,
3961
3962                       open(my $fh, "<", $file)
3963                           || die "Can't open $file: $!";
3964
3965                   otherwise it's necessary to protect any leading and
3966                   trailing whitespace:
3967
3968                       $file =~ s#^(\s)#./$1#;
3969                       open(my $fh, "< $file\0")
3970                           || die "Can't open $file: $!";
3971
3972                   (this may not work on some bizarre filesystems).  One
3973                   should conscientiously choose between the magic and three-
3974                   argument form of "open":
3975
3976                       open(my $in, $ARGV[0]) || die "Can't open $ARGV[0]: $!";
3977
3978                   will allow the user to specify an argument of the form "rsh
3979                   cat file |", but will not work on a filename that happens
3980                   to have a trailing space, while
3981
3982                       open(my $in, "<", $ARGV[0])
3983                           || die "Can't open $ARGV[0]: $!";
3984
3985                   will have exactly the opposite restrictions. (However, some
3986                   shells support the syntax "perl your_program.pl <( rsh cat
3987                   file )", which produces a filename that can be opened
3988                   normally.)
3989
3990               Invoking C-style "open"
3991                   If you want a "real" C open(2), then you should use the
3992                   "sysopen" function, which involves no such magic (but uses
3993                   different filemodes than Perl "open", which corresponds to
3994                   C fopen(3)).  This is another way to protect your filenames
3995                   from interpretation.  For example:
3996
3997                       use IO::Handle;
3998                       sysopen(my $fh, $path, O_RDWR|O_CREAT|O_EXCL)
3999                           or die "Can't open $path: $!";
4000                       $fh->autoflush(1);
4001                       print $fh "stuff $$\n";
4002                       seek($fh, 0, 0);
4003                       print "File contains: ", readline($fh);
4004
4005                   See "seek" for some details about mixing reading and
4006                   writing.
4007
4008               Portability issues
4009                   See "open" in perlport.
4010
4011       opendir DIRHANDLE,EXPR
4012           Opens a directory named EXPR for processing by "readdir",
4013           "telldir", "seekdir", "rewinddir", and "closedir".  Returns true if
4014           successful.  DIRHANDLE may be an expression whose value can be used
4015           as an indirect dirhandle, usually the real dirhandle name.  If
4016           DIRHANDLE is an undefined scalar variable (or array or hash
4017           element), the variable is assigned a reference to a new anonymous
4018           dirhandle; that is, it's autovivified.  Dirhandles are the same
4019           objects as filehandles; an I/O object can only be open as one of
4020           these handle types at once.
4021
4022           See the example at "readdir".
4023
4024       ord EXPR
4025       ord Returns the numeric value of the first character of EXPR.  If EXPR
4026           is an empty string, returns 0.  If EXPR is omitted, uses $_.  (Note
4027           character, not byte.)
4028
4029           For the reverse, see "chr".  See perlunicode for more about
4030           Unicode.
4031
4032       our VARLIST
4033       our TYPE VARLIST
4034       our VARLIST : ATTRS
4035       our TYPE VARLIST : ATTRS
4036           "our" makes a lexical alias to a package (i.e. global) variable of
4037           the same name in the current package for use within the current
4038           lexical scope.
4039
4040           "our" has the same scoping rules as "my" or "state", meaning that
4041           it is only valid within a lexical scope.  Unlike "my" and "state",
4042           which both declare new (lexical) variables, "our" only creates an
4043           alias to an existing variable: a package variable of the same name.
4044
4045           This means that when "use strict 'vars'" is in effect, "our" lets
4046           you use a package variable without qualifying it with the package
4047           name, but only within the lexical scope of the "our" declaration.
4048           This applies immediately--even within the same statement.
4049
4050               package Foo;
4051               use strict;
4052
4053               $Foo::foo = 23;
4054
4055               {
4056                   our $foo;   # alias to $Foo::foo
4057                   print $foo; # prints 23
4058               }
4059
4060               print $Foo::foo; # prints 23
4061
4062               print $foo; # ERROR: requires explicit package name
4063
4064           This works even if the package variable has not been used before,
4065           as package variables spring into existence when first used.
4066
4067               package Foo;
4068               use strict;
4069
4070               our $foo = 23;   # just like $Foo::foo = 23
4071
4072               print $Foo::foo; # prints 23
4073
4074           Because the variable becomes legal immediately under "use strict
4075           'vars'", so long as there is no variable with that name is already
4076           in scope, you can then reference the package variable again even
4077           within the same statement.
4078
4079               package Foo;
4080               use strict;
4081
4082               my  $foo = $foo; # error, undeclared $foo on right-hand side
4083               our $foo = $foo; # no errors
4084
4085           If more than one variable is listed, the list must be placed in
4086           parentheses.
4087
4088               our($bar, $baz);
4089
4090           An "our" declaration declares an alias for a package variable that
4091           will be visible across its entire lexical scope, even across
4092           package boundaries.  The package in which the variable is entered
4093           is determined at the point of the declaration, not at the point of
4094           use.  This means the following behavior holds:
4095
4096               package Foo;
4097               our $bar;      # declares $Foo::bar for rest of lexical scope
4098               $bar = 20;
4099
4100               package Bar;
4101               print $bar;    # prints 20, as it refers to $Foo::bar
4102
4103           Multiple "our" declarations with the same name in the same lexical
4104           scope are allowed if they are in different packages.  If they
4105           happen to be in the same package, Perl will emit warnings if you
4106           have asked for them, just like multiple "my" declarations.  Unlike
4107           a second "my" declaration, which will bind the name to a fresh
4108           variable, a second "our" declaration in the same package, in the
4109           same scope, is merely redundant.
4110
4111               use warnings;
4112               package Foo;
4113               our $bar;      # declares $Foo::bar for rest of lexical scope
4114               $bar = 20;
4115
4116               package Bar;
4117               our $bar = 30; # declares $Bar::bar for rest of lexical scope
4118               print $bar;    # prints 30
4119
4120               our $bar;      # emits warning but has no other effect
4121               print $bar;    # still prints 30
4122
4123           An "our" declaration may also have a list of attributes associated
4124           with it.
4125
4126           The exact semantics and interface of TYPE and ATTRS are still
4127           evolving.  TYPE is currently bound to the use of the fields pragma,
4128           and attributes are handled using the attributes pragma, or,
4129           starting from Perl 5.8.0, also via the Attribute::Handlers module.
4130           See "Private Variables via my()" in perlsub for details.
4131
4132           Note that with a parenthesised list, "undef" can be used as a dummy
4133           placeholder, for example to skip assignment of initial values:
4134
4135               our ( undef, $min, $hour ) = localtime;
4136
4137           "our" differs from "use vars", which allows use of an unqualified
4138           name only within the affected package, but across scopes.
4139
4140       pack TEMPLATE,LIST
4141           Takes a LIST of values and converts it into a string using the
4142           rules given by the TEMPLATE.  The resulting string is the
4143           concatenation of the converted values.  Typically, each converted
4144           value looks like its machine-level representation.  For example, on
4145           32-bit machines an integer may be represented by a sequence of 4
4146           bytes, which  will in Perl be presented as a string that's 4
4147           characters long.
4148
4149           See perlpacktut for an introduction to this function.
4150
4151           The TEMPLATE is a sequence of characters that give the order and
4152           type of values, as follows:
4153
4154               a  A string with arbitrary binary data, will be null padded.
4155               A  A text (ASCII) string, will be space padded.
4156               Z  A null-terminated (ASCIZ) string, will be null padded.
4157
4158               b  A bit string (ascending bit order inside each byte,
4159                  like vec()).
4160               B  A bit string (descending bit order inside each byte).
4161               h  A hex string (low nybble first).
4162               H  A hex string (high nybble first).
4163
4164               c  A signed char (8-bit) value.
4165               C  An unsigned char (octet) value.
4166               W  An unsigned char value (can be greater than 255).
4167
4168               s  A signed short (16-bit) value.
4169               S  An unsigned short value.
4170
4171               l  A signed long (32-bit) value.
4172               L  An unsigned long value.
4173
4174               q  A signed quad (64-bit) value.
4175               Q  An unsigned quad value.
4176                    (Quads are available only if your system supports 64-bit
4177                     integer values _and_ if Perl has been compiled to support
4178                     those.  Raises an exception otherwise.)
4179
4180               i  A signed integer value.
4181               I  An unsigned integer value.
4182                    (This 'integer' is _at_least_ 32 bits wide.  Its exact
4183                     size depends on what a local C compiler calls 'int'.)
4184
4185               n  An unsigned short (16-bit) in "network" (big-endian) order.
4186               N  An unsigned long (32-bit) in "network" (big-endian) order.
4187               v  An unsigned short (16-bit) in "VAX" (little-endian) order.
4188               V  An unsigned long (32-bit) in "VAX" (little-endian) order.
4189
4190               j  A Perl internal signed integer value (IV).
4191               J  A Perl internal unsigned integer value (UV).
4192
4193               f  A single-precision float in native format.
4194               d  A double-precision float in native format.
4195
4196               F  A Perl internal floating-point value (NV) in native format
4197               D  A float of long-double precision in native format.
4198                    (Long doubles are available only if your system supports
4199                     long double values _and_ if Perl has been compiled to
4200                     support those.  Raises an exception otherwise.
4201                     Note that there are different long double formats.)
4202
4203               p  A pointer to a null-terminated string.
4204               P  A pointer to a structure (fixed-length string).
4205
4206               u  A uuencoded string.
4207               U  A Unicode character number.  Encodes to a character in char-
4208                  acter mode and UTF-8 (or UTF-EBCDIC in EBCDIC platforms) in
4209                  byte mode.
4210
4211               w  A BER compressed integer (not an ASN.1 BER, see perlpacktut
4212                  for details).  Its bytes represent an unsigned integer in
4213                  base 128, most significant digit first, with as few digits
4214                  as possible.  Bit eight (the high bit) is set on each byte
4215                  except the last.
4216
4217               x  A null byte (a.k.a ASCII NUL, "\000", chr(0))
4218               X  Back up a byte.
4219               @  Null-fill or truncate to absolute position, counted from the
4220                  start of the innermost ()-group.
4221               .  Null-fill or truncate to absolute position specified by
4222                  the value.
4223               (  Start of a ()-group.
4224
4225           One or more modifiers below may optionally follow certain letters
4226           in the TEMPLATE (the second column lists letters for which the
4227           modifier is valid):
4228
4229               !   sSlLiI     Forces native (short, long, int) sizes instead
4230                              of fixed (16-/32-bit) sizes.
4231
4232               !   xX         Make x and X act as alignment commands.
4233
4234               !   nNvV       Treat integers as signed instead of unsigned.
4235
4236               !   @.         Specify position as byte offset in the internal
4237                              representation of the packed string.  Efficient
4238                              but dangerous.
4239
4240               >   sSiIlLqQ   Force big-endian byte-order on the type.
4241                   jJfFdDpP   (The "big end" touches the construct.)
4242
4243               <   sSiIlLqQ   Force little-endian byte-order on the type.
4244                   jJfFdDpP   (The "little end" touches the construct.)
4245
4246           The ">" and "<" modifiers can also be used on "()" groups to force
4247           a particular byte-order on all components in that group, including
4248           all its subgroups.
4249
4250           The following rules apply:
4251
4252           •   Each letter may optionally be followed by a number indicating
4253               the repeat count.  A numeric repeat count may optionally be
4254               enclosed in brackets, as in "pack("C[80]", @arr)".  The repeat
4255               count gobbles that many values from the LIST when used with all
4256               format types other than "a", "A", "Z", "b", "B", "h", "H", "@",
4257               ".", "x", "X", and "P", where it means something else,
4258               described below.  Supplying a "*" for the repeat count instead
4259               of a number means to use however many items are left, except
4260               for:
4261
4262               •   "@", "x", and "X", where it is equivalent to 0.
4263
4264               •   <.>, where it means relative to the start of the string.
4265
4266               •   "u", where it is equivalent to 1 (or 45, which here is
4267                   equivalent).
4268
4269               One can replace a numeric repeat count with a template letter
4270               enclosed in brackets to use the packed byte length of the
4271               bracketed template for the repeat count.
4272
4273               For example, the template "x[L]" skips as many bytes as in a
4274               packed long, and the template "$t X[$t] $t" unpacks twice
4275               whatever $t (when variable-expanded) unpacks.  If the template
4276               in brackets contains alignment commands (such as "x![d]"), its
4277               packed length is calculated as if the start of the template had
4278               the maximal possible alignment.
4279
4280               When used with "Z", a "*" as the repeat count is guaranteed to
4281               add a trailing null byte, so the resulting string is always one
4282               byte longer than the byte length of the item itself.
4283
4284               When used with "@", the repeat count represents an offset from
4285               the start of the innermost "()" group.
4286
4287               When used with ".", the repeat count determines the starting
4288               position to calculate the value offset as follows:
4289
4290               •   If the repeat count is 0, it's relative to the current
4291                   position.
4292
4293               •   If the repeat count is "*", the offset is relative to the
4294                   start of the packed string.
4295
4296               •   And if it's an integer n, the offset is relative to the
4297                   start of the nth innermost "( )" group, or to the start of
4298                   the string if n is bigger then the group level.
4299
4300               The repeat count for "u" is interpreted as the maximal number
4301               of bytes to encode per line of output, with 0, 1 and 2 replaced
4302               by 45.  The repeat count should not be more than 65.
4303
4304           •   The "a", "A", and "Z" types gobble just one value, but pack it
4305               as a string of length count, padding with nulls or spaces as
4306               needed.  When unpacking, "A" strips trailing whitespace and
4307               nulls, "Z" strips everything after the first null, and "a"
4308               returns data with no stripping at all.
4309
4310               If the value to pack is too long, the result is truncated.  If
4311               it's too long and an explicit count is provided, "Z" packs only
4312               "$count-1" bytes, followed by a null byte.  Thus "Z" always
4313               packs a trailing null, except when the count is 0.
4314
4315           •   Likewise, the "b" and "B" formats pack a string that's that
4316               many bits long.  Each such format generates 1 bit of the
4317               result.  These are typically followed by a repeat count like
4318               "B8" or "B64".
4319
4320               Each result bit is based on the least-significant bit of the
4321               corresponding input character, i.e., on "ord($char)%2".  In
4322               particular, characters "0" and "1" generate bits 0 and 1, as do
4323               characters "\000" and "\001".
4324
4325               Starting from the beginning of the input string, each 8-tuple
4326               of characters is converted to 1 character of output.  With
4327               format "b", the first character of the 8-tuple determines the
4328               least-significant bit of a character; with format "B", it
4329               determines the most-significant bit of a character.
4330
4331               If the length of the input string is not evenly divisible by 8,
4332               the remainder is packed as if the input string were padded by
4333               null characters at the end.  Similarly during unpacking,
4334               "extra" bits are ignored.
4335
4336               If the input string is longer than needed, remaining characters
4337               are ignored.
4338
4339               A "*" for the repeat count uses all characters of the input
4340               field.  On unpacking, bits are converted to a string of 0s and
4341               1s.
4342
4343           •   The "h" and "H" formats pack a string that many nybbles (4-bit
4344               groups, representable as hexadecimal digits, "0".."9" "a".."f")
4345               long.
4346
4347               For each such format, "pack" generates 4 bits of result.  With
4348               non-alphabetical characters, the result is based on the 4
4349               least-significant bits of the input character, i.e., on
4350               "ord($char)%16".  In particular, characters "0" and "1"
4351               generate nybbles 0 and 1, as do bytes "\000" and "\001".  For
4352               characters "a".."f" and "A".."F", the result is compatible with
4353               the usual hexadecimal digits, so that "a" and "A" both generate
4354               the nybble "0xA==10".  Use only these specific hex characters
4355               with this format.
4356
4357               Starting from the beginning of the template to "pack", each
4358               pair of characters is converted to 1 character of output.  With
4359               format "h", the first character of the pair determines the
4360               least-significant nybble of the output character; with format
4361               "H", it determines the most-significant nybble.
4362
4363               If the length of the input string is not even, it behaves as if
4364               padded by a null character at the end.  Similarly, "extra"
4365               nybbles are ignored during unpacking.
4366
4367               If the input string is longer than needed, extra characters are
4368               ignored.
4369
4370               A "*" for the repeat count uses all characters of the input
4371               field.  For "unpack", nybbles are converted to a string of
4372               hexadecimal digits.
4373
4374           •   The "p" format packs a pointer to a null-terminated string.
4375               You are responsible for ensuring that the string is not a
4376               temporary value, as that could potentially get deallocated
4377               before you got around to using the packed result.  The "P"
4378               format packs a pointer to a structure of the size indicated by
4379               the length.  A null pointer is created if the corresponding
4380               value for "p" or "P" is "undef"; similarly with "unpack", where
4381               a null pointer unpacks into "undef".
4382
4383               If your system has a strange pointer size--meaning a pointer is
4384               neither as big as an int nor as big as a long--it may not be
4385               possible to pack or unpack pointers in big- or little-endian
4386               byte order.  Attempting to do so raises an exception.
4387
4388           •   The "/" template character allows packing and unpacking of a
4389               sequence of items where the packed structure contains a packed
4390               item count followed by the packed items themselves.  This is
4391               useful when the structure you're unpacking has encoded the
4392               sizes or repeat counts for some of its fields within the
4393               structure itself as separate fields.
4394
4395               For "pack", you write length-item"/"sequence-item, and the
4396               length-item describes how the length value is packed.  Formats
4397               likely to be of most use are integer-packing ones like "n" for
4398               Java strings, "w" for ASN.1 or SNMP, and "N" for Sun XDR.
4399
4400               For "pack", sequence-item may have a repeat count, in which
4401               case the minimum of that and the number of available items is
4402               used as the argument for length-item.  If it has no repeat
4403               count or uses a '*', the number of available items is used.
4404
4405               For "unpack", an internal stack of integer arguments unpacked
4406               so far is used.  You write "/"sequence-item and the repeat
4407               count is obtained by popping off the last element from the
4408               stack.  The sequence-item must not have a repeat count.
4409
4410               If sequence-item refers to a string type ("A", "a", or "Z"),
4411               the length-item is the string length, not the number of
4412               strings.  With an explicit repeat count for pack, the packed
4413               string is adjusted to that length.  For example:
4414
4415                This code:                             gives this result:
4416
4417                unpack("W/a", "\004Gurusamy")          ("Guru")
4418                unpack("a3/A A*", "007 Bond  J ")      (" Bond", "J")
4419                unpack("a3 x2 /A A*", "007: Bond, J.") ("Bond, J", ".")
4420
4421                pack("n/a* w/a","hello,","world")     "\000\006hello,\005world"
4422                pack("a/W2", ord("a") .. ord("z"))    "2ab"
4423
4424               The length-item is not returned explicitly from "unpack".
4425
4426               Supplying a count to the length-item format letter is only
4427               useful with "A", "a", or "Z".  Packing with a length-item of
4428               "a" or "Z" may introduce "\000" characters, which Perl does not
4429               regard as legal in numeric strings.
4430
4431           •   The integer types "s", "S", "l", and "L" may be followed by a
4432               "!" modifier to specify native shorts or longs.  As shown in
4433               the example above, a bare "l" means exactly 32 bits, although
4434               the native "long" as seen by the local C compiler may be
4435               larger.  This is mainly an issue on 64-bit platforms.  You can
4436               see whether using "!" makes any difference this way:
4437
4438                   printf "format s is %d, s! is %d\n",
4439                       length pack("s"), length pack("s!");
4440
4441                   printf "format l is %d, l! is %d\n",
4442                       length pack("l"), length pack("l!");
4443
4444               "i!" and "I!" are also allowed, but only for completeness'
4445               sake: they are identical to "i" and "I".
4446
4447               The actual sizes (in bytes) of native shorts, ints, longs, and
4448               long longs on the platform where Perl was built are also
4449               available from the command line:
4450
4451                   $ perl -V:{short,int,long{,long}}size
4452                   shortsize='2';
4453                   intsize='4';
4454                   longsize='4';
4455                   longlongsize='8';
4456
4457               or programmatically via the "Config" module:
4458
4459                      use Config;
4460                      print $Config{shortsize},    "\n";
4461                      print $Config{intsize},      "\n";
4462                      print $Config{longsize},     "\n";
4463                      print $Config{longlongsize}, "\n";
4464
4465               $Config{longlongsize} is undefined on systems without long long
4466               support.
4467
4468           •   The integer formats "s", "S", "i", "I", "l", "L", "j", and "J"
4469               are inherently non-portable between processors and operating
4470               systems because they obey native byteorder and endianness.  For
4471               example, a 4-byte integer 0x12345678 (305419896 decimal) would
4472               be ordered natively (arranged in and handled by the CPU
4473               registers) into bytes as
4474
4475                   0x12 0x34 0x56 0x78  # big-endian
4476                   0x78 0x56 0x34 0x12  # little-endian
4477
4478               Basically, Intel and VAX CPUs are little-endian, while
4479               everybody else, including Motorola m68k/88k, PPC, Sparc, HP PA,
4480               Power, and Cray, are big-endian.  Alpha and MIPS can be either:
4481               Digital/Compaq uses (well, used) them in little-endian mode,
4482               but SGI/Cray uses them in big-endian mode.
4483
4484               The names big-endian and little-endian are comic references to
4485               the egg-eating habits of the little-endian Lilliputians and the
4486               big-endian Blefuscudians from the classic Jonathan Swift
4487               satire, Gulliver's Travels.  This entered computer lingo via
4488               the paper "On Holy Wars and a Plea for Peace" by Danny Cohen,
4489               USC/ISI IEN 137, April 1, 1980.
4490
4491               Some systems may have even weirder byte orders such as
4492
4493                  0x56 0x78 0x12 0x34
4494                  0x34 0x12 0x78 0x56
4495
4496               These are called mid-endian, middle-endian, mixed-endian, or
4497               just weird.
4498
4499               You can determine your system endianness with this incantation:
4500
4501                  printf("%#02x ", $_) for unpack("W*", pack L=>0x12345678);
4502
4503               The byteorder on the platform where Perl was built is also
4504               available via Config:
4505
4506                   use Config;
4507                   print "$Config{byteorder}\n";
4508
4509               or from the command line:
4510
4511                   $ perl -V:byteorder
4512
4513               Byteorders "1234" and "12345678" are little-endian; "4321" and
4514               "87654321" are big-endian.  Systems with multiarchitecture
4515               binaries will have "ffff", signifying that static information
4516               doesn't work, one must use runtime probing.
4517
4518               For portably packed integers, either use the formats "n", "N",
4519               "v", and "V" or else use the ">" and "<" modifiers described
4520               immediately below.  See also perlport.
4521
4522           •   Also floating point numbers have endianness.  Usually (but not
4523               always) this agrees with the integer endianness.  Even though
4524               most platforms these days use the IEEE 754 binary format, there
4525               are differences, especially if the long doubles are involved.
4526               You can see the "Config" variables "doublekind" and
4527               "longdblkind" (also "doublesize", "longdblsize"): the "kind"
4528               values are enums, unlike "byteorder".
4529
4530               Portability-wise the best option is probably to keep to the
4531               IEEE 754 64-bit doubles, and of agreed-upon endianness.
4532               Another possibility is the "%a") format of "printf".
4533
4534           •   Starting with Perl 5.10.0, integer and floating-point formats,
4535               along with the "p" and "P" formats and "()" groups, may all be
4536               followed by the ">" or "<" endianness modifiers to respectively
4537               enforce big- or little-endian byte-order.  These modifiers are
4538               especially useful given how "n", "N", "v", and "V" don't cover
4539               signed integers, 64-bit integers, or floating-point values.
4540
4541               Here are some concerns to keep in mind when using an endianness
4542               modifier:
4543
4544               •   Exchanging signed integers between different platforms
4545                   works only when all platforms store them in the same
4546                   format.  Most platforms store signed integers in two's-
4547                   complement notation, so usually this is not an issue.
4548
4549               •   The ">" or "<" modifiers can only be used on floating-point
4550                   formats on big- or little-endian machines.  Otherwise,
4551                   attempting to use them raises an exception.
4552
4553               •   Forcing big- or little-endian byte-order on floating-point
4554                   values for data exchange can work only if all platforms use
4555                   the same binary representation such as IEEE floating-point.
4556                   Even if all platforms are using IEEE, there may still be
4557                   subtle differences.  Being able to use ">" or "<" on
4558                   floating-point values can be useful, but also dangerous if
4559                   you don't know exactly what you're doing.  It is not a
4560                   general way to portably store floating-point values.
4561
4562               •   When using ">" or "<" on a "()" group, this affects all
4563                   types inside the group that accept byte-order modifiers,
4564                   including all subgroups.  It is silently ignored for all
4565                   other types.  You are not allowed to override the byte-
4566                   order within a group that already has a byte-order modifier
4567                   suffix.
4568
4569           •   Real numbers (floats and doubles) are in native machine format
4570               only.  Due to the multiplicity of floating-point formats and
4571               the lack of a standard "network" representation for them, no
4572               facility for interchange has been made.  This means that packed
4573               floating-point data written on one machine may not be readable
4574               on another, even if both use IEEE floating-point arithmetic
4575               (because the endianness of the memory representation is not
4576               part of the IEEE spec).  See also perlport.
4577
4578               If you know exactly what you're doing, you can use the ">" or
4579               "<" modifiers to force big- or little-endian byte-order on
4580               floating-point values.
4581
4582               Because Perl uses doubles (or long doubles, if configured)
4583               internally for all numeric calculation, converting from double
4584               into float and thence to double again loses precision, so
4585               "unpack("f", pack("f", $foo)") will not in general equal $foo.
4586
4587           •   Pack and unpack can operate in two modes: character mode ("C0"
4588               mode) where the packed string is processed per character, and
4589               UTF-8 byte mode ("U0" mode) where the packed string is
4590               processed in its UTF-8-encoded Unicode form on a byte-by-byte
4591               basis.  Character mode is the default unless the format string
4592               starts with "U".  You can always switch mode mid-format with an
4593               explicit "C0" or "U0" in the format.  This mode remains in
4594               effect until the next mode change, or until the end of the "()"
4595               group it (directly) applies to.
4596
4597               Using "C0" to get Unicode characters while using "U0" to get
4598               non-Unicode bytes is not necessarily obvious.   Probably only
4599               the first of these is what you want:
4600
4601                   $ perl -CS -E 'say "\x{3B1}\x{3C9}"' |
4602                     perl -CS -ne 'printf "%v04X\n", $_ for unpack("C0A*", $_)'
4603                   03B1.03C9
4604                   $ perl -CS -E 'say "\x{3B1}\x{3C9}"' |
4605                     perl -CS -ne 'printf "%v02X\n", $_ for unpack("U0A*", $_)'
4606                   CE.B1.CF.89
4607                   $ perl -CS -E 'say "\x{3B1}\x{3C9}"' |
4608                     perl -C0 -ne 'printf "%v02X\n", $_ for unpack("C0A*", $_)'
4609                   CE.B1.CF.89
4610                   $ perl -CS -E 'say "\x{3B1}\x{3C9}"' |
4611                     perl -C0 -ne 'printf "%v02X\n", $_ for unpack("U0A*", $_)'
4612                   C3.8E.C2.B1.C3.8F.C2.89
4613
4614               Those examples also illustrate that you should not try to use
4615               "pack"/"unpack" as a substitute for the Encode module.
4616
4617           •   You must yourself do any alignment or padding by inserting, for
4618               example, enough "x"es while packing.  There is no way for
4619               "pack" and "unpack" to know where characters are going to or
4620               coming from, so they handle their output and input as flat
4621               sequences of characters.
4622
4623           •   A "()" group is a sub-TEMPLATE enclosed in parentheses.  A
4624               group may take a repeat count either as postfix, or for
4625               "unpack", also via the "/" template character.  Within each
4626               repetition of a group, positioning with "@" starts over at 0.
4627               Therefore, the result of
4628
4629                   pack("@1A((@2A)@3A)", qw[X Y Z])
4630
4631               is the string "\0X\0\0YZ".
4632
4633           •   "x" and "X" accept the "!" modifier to act as alignment
4634               commands: they jump forward or back to the closest position
4635               aligned at a multiple of "count" characters.  For example, to
4636               "pack" or "unpack" a C structure like
4637
4638                   struct {
4639                       char   c;    /* one signed, 8-bit character */
4640                       double d;
4641                       char   cc[2];
4642                   }
4643
4644               one may need to use the template "c x![d] d c[2]".  This
4645               assumes that doubles must be aligned to the size of double.
4646
4647               For alignment commands, a "count" of 0 is equivalent to a
4648               "count" of 1; both are no-ops.
4649
4650           •   "n", "N", "v" and "V" accept the "!" modifier to represent
4651               signed 16-/32-bit integers in big-/little-endian order.  This
4652               is portable only when all platforms sharing packed data use the
4653               same binary representation for signed integers; for example,
4654               when all platforms use two's-complement representation.
4655
4656           •   Comments can be embedded in a TEMPLATE using "#" through the
4657               end of line.  White space can separate pack codes from each
4658               other, but modifiers and repeat counts must follow immediately.
4659               Breaking complex templates into individual line-by-line
4660               components, suitably annotated, can do as much to improve
4661               legibility and maintainability of pack/unpack formats as "/x"
4662               can for complicated pattern matches.
4663
4664           •   If TEMPLATE requires more arguments than "pack" is given,
4665               "pack" assumes additional "" arguments.  If TEMPLATE requires
4666               fewer arguments than given, extra arguments are ignored.
4667
4668           •   Attempting to pack the special floating point values "Inf" and
4669               "NaN" (infinity, also in negative, and not-a-number) into
4670               packed integer values (like "L") is a fatal error.  The reason
4671               for this is that there simply isn't any sensible mapping for
4672               these special values into integers.
4673
4674           Examples:
4675
4676               $foo = pack("WWWW",65,66,67,68);
4677               # foo eq "ABCD"
4678               $foo = pack("W4",65,66,67,68);
4679               # same thing
4680               $foo = pack("W4",0x24b6,0x24b7,0x24b8,0x24b9);
4681               # same thing with Unicode circled letters.
4682               $foo = pack("U4",0x24b6,0x24b7,0x24b8,0x24b9);
4683               # same thing with Unicode circled letters.  You don't get the
4684               # UTF-8 bytes because the U at the start of the format caused
4685               # a switch to U0-mode, so the UTF-8 bytes get joined into
4686               # characters
4687               $foo = pack("C0U4",0x24b6,0x24b7,0x24b8,0x24b9);
4688               # foo eq "\xe2\x92\xb6\xe2\x92\xb7\xe2\x92\xb8\xe2\x92\xb9"
4689               # This is the UTF-8 encoding of the string in the
4690               # previous example
4691
4692               $foo = pack("ccxxcc",65,66,67,68);
4693               # foo eq "AB\0\0CD"
4694
4695               # NOTE: The examples above featuring "W" and "c" are true
4696               # only on ASCII and ASCII-derived systems such as ISO Latin 1
4697               # and UTF-8.  On EBCDIC systems, the first example would be
4698               #      $foo = pack("WWWW",193,194,195,196);
4699
4700               $foo = pack("s2",1,2);
4701               # "\001\000\002\000" on little-endian
4702               # "\000\001\000\002" on big-endian
4703
4704               $foo = pack("a4","abcd","x","y","z");
4705               # "abcd"
4706
4707               $foo = pack("aaaa","abcd","x","y","z");
4708               # "axyz"
4709
4710               $foo = pack("a14","abcdefg");
4711               # "abcdefg\0\0\0\0\0\0\0"
4712
4713               $foo = pack("i9pl", gmtime);
4714               # a real struct tm (on my system anyway)
4715
4716               $utmp_template = "Z8 Z8 Z16 L";
4717               $utmp = pack($utmp_template, @utmp1);
4718               # a struct utmp (BSDish)
4719
4720               @utmp2 = unpack($utmp_template, $utmp);
4721               # "@utmp1" eq "@utmp2"
4722
4723               sub bintodec {
4724                   unpack("N", pack("B32", substr("0" x 32 . shift, -32)));
4725               }
4726
4727               $foo = pack('sx2l', 12, 34);
4728               # short 12, two zero bytes padding, long 34
4729               $bar = pack('s@4l', 12, 34);
4730               # short 12, zero fill to position 4, long 34
4731               # $foo eq $bar
4732               $baz = pack('s.l', 12, 4, 34);
4733               # short 12, zero fill to position 4, long 34
4734
4735               $foo = pack('nN', 42, 4711);
4736               # pack big-endian 16- and 32-bit unsigned integers
4737               $foo = pack('S>L>', 42, 4711);
4738               # exactly the same
4739               $foo = pack('s<l<', -42, 4711);
4740               # pack little-endian 16- and 32-bit signed integers
4741               $foo = pack('(sl)<', -42, 4711);
4742               # exactly the same
4743
4744           The same template may generally also be used in "unpack".
4745
4746       package NAMESPACE
4747       package NAMESPACE VERSION
4748       package NAMESPACE BLOCK
4749       package NAMESPACE VERSION BLOCK
4750           Declares the BLOCK or the rest of the compilation unit as being in
4751           the given namespace.  The scope of the package declaration is
4752           either the supplied code BLOCK or, in the absence of a BLOCK, from
4753           the declaration itself through the end of current scope (the
4754           enclosing block, file, or "eval").  That is, the forms without a
4755           BLOCK are operative through the end of the current scope, just like
4756           the "my", "state", and "our" operators.  All unqualified dynamic
4757           identifiers in this scope will be in the given namespace, except
4758           where overridden by another "package" declaration or when they're
4759           one of the special identifiers that qualify into "main::", like
4760           "STDOUT", "ARGV", "ENV", and the punctuation variables.
4761
4762           A package statement affects dynamic variables only, including those
4763           you've used "local" on, but not lexically-scoped variables, which
4764           are created with "my", "state", or "our".  Typically it would be
4765           the first declaration in a file included by "require" or "use".
4766           You can switch into a package in more than one place, since this
4767           only determines which default symbol table the compiler uses for
4768           the rest of that block.  You can refer to identifiers in other
4769           packages than the current one by prefixing the identifier with the
4770           package name and a double colon, as in $SomePack::var or
4771           "ThatPack::INPUT_HANDLE".  If package name is omitted, the "main"
4772           package as assumed.  That is, $::sail is equivalent to $main::sail
4773           (as well as to "$main'sail", still seen in ancient code, mostly
4774           from Perl 4).
4775
4776           If VERSION is provided, "package" sets the $VERSION variable in the
4777           given namespace to a version object with the VERSION provided.
4778           VERSION must be a "strict" style version number as defined by the
4779           version module: a positive decimal number (integer or decimal-
4780           fraction) without exponentiation or else a dotted-decimal v-string
4781           with a leading 'v' character and at least three components.  You
4782           should set $VERSION only once per package.
4783
4784           See "Packages" in perlmod for more information about packages,
4785           modules, and classes.  See perlsub for other scoping issues.
4786
4787       __PACKAGE__
4788           A special token that returns the name of the package in which it
4789           occurs.
4790
4791       pipe READHANDLE,WRITEHANDLE
4792           Opens a pair of connected pipes like the corresponding system call.
4793           Note that if you set up a loop of piped processes, deadlock can
4794           occur unless you are very careful.  In addition, note that Perl's
4795           pipes use IO buffering, so you may need to set $| to flush your
4796           WRITEHANDLE after each command, depending on the application.
4797
4798           Returns true on success.
4799
4800           See IPC::Open2, IPC::Open3, and "Bidirectional Communication with
4801           Another Process" in perlipc for examples of such things.
4802
4803           On systems that support a close-on-exec flag on files, that flag is
4804           set on all newly opened file descriptors whose "fileno"s are higher
4805           than the current value of $^F (by default 2 for "STDERR").  See
4806           "$^F" in perlvar.
4807
4808       pop ARRAY
4809       pop Pops and returns the last value of the array, shortening the array
4810           by one element.
4811
4812           Returns the undefined value if the array is empty, although this
4813           may also happen at other times.  If ARRAY is omitted, pops the
4814           @ARGV array in the main program, but the @_ array in subroutines,
4815           just like "shift".
4816
4817           Starting with Perl 5.14, an experimental feature allowed "pop" to
4818           take a scalar expression. This experiment has been deemed
4819           unsuccessful, and was removed as of Perl 5.24.
4820
4821       pos SCALAR
4822       pos Returns the offset of where the last "m//g" search left off for the
4823           variable in question ($_ is used when the variable is not
4824           specified).  This offset is in characters unless the (no-longer-
4825           recommended) "use bytes" pragma is in effect, in which case the
4826           offset is in bytes.  Note that 0 is a valid match offset.  "undef"
4827           indicates that the search position is reset (usually due to match
4828           failure, but can also be because no match has yet been run on the
4829           scalar).
4830
4831           "pos" directly accesses the location used by the regexp engine to
4832           store the offset, so assigning to "pos" will change that offset,
4833           and so will also influence the "\G" zero-width assertion in regular
4834           expressions.  Both of these effects take place for the next match,
4835           so you can't affect the position with "pos" during the current
4836           match, such as in "(?{pos() = 5})" or "s//pos() = 5/e".
4837
4838           Setting "pos" also resets the matched with zero-length flag,
4839           described under "Repeated Patterns Matching a Zero-length
4840           Substring" in perlre.
4841
4842           Because a failed "m//gc" match doesn't reset the offset, the return
4843           from "pos" won't change either in this case.  See perlre and
4844           perlop.
4845
4846       print FILEHANDLE LIST
4847       print FILEHANDLE
4848       print LIST
4849       print
4850           Prints a string or a list of strings.  Returns true if successful.
4851           FILEHANDLE may be a scalar variable containing the name of or a
4852           reference to the filehandle, thus introducing one level of
4853           indirection.  (NOTE: If FILEHANDLE is a variable and the next token
4854           is a term, it may be misinterpreted as an operator unless you
4855           interpose a "+" or put parentheses around the arguments.)  If
4856           FILEHANDLE is omitted, prints to the last selected (see "select")
4857           output handle.  If LIST is omitted, prints $_ to the currently
4858           selected output handle.  To use FILEHANDLE alone to print the
4859           content of $_ to it, you must use a bareword filehandle like "FH",
4860           not an indirect one like $fh.  To set the default output handle to
4861           something other than STDOUT, use the select operation.
4862
4863           The current value of $, (if any) is printed between each LIST item.
4864           The current value of "$\" (if any) is printed after the entire LIST
4865           has been printed.  Because print takes a LIST, anything in the LIST
4866           is evaluated in list context, including any subroutines whose
4867           return lists you pass to "print".  Be careful not to follow the
4868           print keyword with a left parenthesis unless you want the
4869           corresponding right parenthesis to terminate the arguments to the
4870           print; put parentheses around all arguments (or interpose a "+",
4871           but that doesn't look as good).
4872
4873           If you're storing handles in an array or hash, or in general
4874           whenever you're using any expression more complex than a bareword
4875           handle or a plain, unsubscripted scalar variable to retrieve it,
4876           you will have to use a block returning the filehandle value
4877           instead, in which case the LIST may not be omitted:
4878
4879               print { $files[$i] } "stuff\n";
4880               print { $OK ? *STDOUT : *STDERR } "stuff\n";
4881
4882           Printing to a closed pipe or socket will generate a SIGPIPE signal.
4883           See perlipc for more on signal handling.
4884
4885       printf FILEHANDLE FORMAT, LIST
4886       printf FILEHANDLE
4887       printf FORMAT, LIST
4888       printf
4889           Equivalent to "print FILEHANDLE sprintf(FORMAT, LIST)", except that
4890           "$\" (the output record separator) is not appended.  The FORMAT and
4891           the LIST are actually parsed as a single list.  The first argument
4892           of the list will be interpreted as the "printf" format.  This means
4893           that "printf(@_)" will use $_[0] as the format.  See sprintf for an
4894           explanation of the format argument.  If "use locale" (including
4895           "use locale ':not_characters'") is in effect and "POSIX::setlocale"
4896           has been called, the character used for the decimal separator in
4897           formatted floating-point numbers is affected by the "LC_NUMERIC"
4898           locale setting.  See perllocale and POSIX.
4899
4900           For historical reasons, if you omit the list, $_ is used as the
4901           format; to use FILEHANDLE without a list, you must use a bareword
4902           filehandle like "FH", not an indirect one like $fh.  However, this
4903           will rarely do what you want; if $_ contains formatting codes, they
4904           will be replaced with the empty string and a warning will be
4905           emitted if warnings are enabled.  Just use "print" if you want to
4906           print the contents of $_.
4907
4908           Don't fall into the trap of using a "printf" when a simple "print"
4909           would do.  The "print" is more efficient and less error prone.
4910
4911       prototype FUNCTION
4912       prototype
4913           Returns the prototype of a function as a string (or "undef" if the
4914           function has no prototype).  FUNCTION is a reference to, or the
4915           name of, the function whose prototype you want to retrieve.  If
4916           FUNCTION is omitted, $_ is used.
4917
4918           If FUNCTION is a string starting with "CORE::", the rest is taken
4919           as a name for a Perl builtin.  If the builtin's arguments cannot be
4920           adequately expressed by a prototype (such as "system"), "prototype"
4921           returns "undef", because the builtin does not really behave like a
4922           Perl function.  Otherwise, the string describing the equivalent
4923           prototype is returned.
4924
4925       push ARRAY,LIST
4926           Treats ARRAY as a stack by appending the values of LIST to the end
4927           of ARRAY.  The length of ARRAY increases by the length of LIST.
4928           Has the same effect as
4929
4930               for my $value (LIST) {
4931                   $ARRAY[++$#ARRAY] = $value;
4932               }
4933
4934           but is more efficient.  Returns the number of elements in the array
4935           following the completed "push".
4936
4937           Starting with Perl 5.14, an experimental feature allowed "push" to
4938           take a scalar expression. This experiment has been deemed
4939           unsuccessful, and was removed as of Perl 5.24.
4940
4941       q/STRING/
4942       qq/STRING/
4943       qw/STRING/
4944       qx/STRING/
4945           Generalized quotes.  See "Quote-Like Operators" in perlop.
4946
4947       qr/STRING/
4948           Regexp-like quote.  See "Regexp Quote-Like Operators" in perlop.
4949
4950       quotemeta EXPR
4951       quotemeta
4952           Returns the value of EXPR with all the ASCII non-"word" characters
4953           backslashed.  (That is, all ASCII characters not matching
4954           "/[A-Za-z_0-9]/" will be preceded by a backslash in the returned
4955           string, regardless of any locale settings.)  This is the internal
4956           function implementing the "\Q" escape in double-quoted strings.
4957           (See below for the behavior on non-ASCII code points.)
4958
4959           If EXPR is omitted, uses $_.
4960
4961           quotemeta (and "\Q" ... "\E") are useful when interpolating strings
4962           into regular expressions, because by default an interpolated
4963           variable will be considered a mini-regular expression.  For
4964           example:
4965
4966               my $sentence = 'The quick brown fox jumped over the lazy dog';
4967               my $substring = 'quick.*?fox';
4968               $sentence =~ s{$substring}{big bad wolf};
4969
4970           Will cause $sentence to become 'The big bad wolf jumped over...'.
4971
4972           On the other hand:
4973
4974               my $sentence = 'The quick brown fox jumped over the lazy dog';
4975               my $substring = 'quick.*?fox';
4976               $sentence =~ s{\Q$substring\E}{big bad wolf};
4977
4978           Or:
4979
4980               my $sentence = 'The quick brown fox jumped over the lazy dog';
4981               my $substring = 'quick.*?fox';
4982               my $quoted_substring = quotemeta($substring);
4983               $sentence =~ s{$quoted_substring}{big bad wolf};
4984
4985           Will both leave the sentence as is.  Normally, when accepting
4986           literal string input from the user, "quotemeta" or "\Q" must be
4987           used.
4988
4989           In Perl v5.14, all non-ASCII characters are quoted in
4990           non-UTF-8-encoded strings, but not quoted in UTF-8 strings.
4991
4992           Starting in Perl v5.16, Perl adopted a Unicode-defined strategy for
4993           quoting non-ASCII characters; the quoting of ASCII characters is
4994           unchanged.
4995
4996           Also unchanged is the quoting of non-UTF-8 strings when outside the
4997           scope of a "use feature 'unicode_strings'", which is to quote all
4998           characters in the upper Latin1 range.  This provides complete
4999           backwards compatibility for old programs which do not use Unicode.
5000           (Note that "unicode_strings" is automatically enabled within the
5001           scope of a "use v5.12" or greater.)
5002
5003           Within the scope of "use locale", all non-ASCII Latin1 code points
5004           are quoted whether the string is encoded as UTF-8 or not.  As
5005           mentioned above, locale does not affect the quoting of ASCII-range
5006           characters.  This protects against those locales where characters
5007           such as "|" are considered to be word characters.
5008
5009           Otherwise, Perl quotes non-ASCII characters using an adaptation
5010           from Unicode (see <https://www.unicode.org/reports/tr31/>).  The
5011           only code points that are quoted are those that have any of the
5012           Unicode properties:  Pattern_Syntax, Pattern_White_Space,
5013           White_Space, Default_Ignorable_Code_Point, or
5014           General_Category=Control.
5015
5016           Of these properties, the two important ones are Pattern_Syntax and
5017           Pattern_White_Space.  They have been set up by Unicode for exactly
5018           this purpose of deciding which characters in a regular expression
5019           pattern should be quoted.  No character that can be in an
5020           identifier has these properties.
5021
5022           Perl promises, that if we ever add regular expression pattern
5023           metacharacters to the dozen already defined ("\ | ( ) [ { ^ $ * + ?
5024           ."), that we will only use ones that have the Pattern_Syntax
5025           property.  Perl also promises, that if we ever add characters that
5026           are considered to be white space in regular expressions (currently
5027           mostly affected by "/x"), they will all have the
5028           Pattern_White_Space property.
5029
5030           Unicode promises that the set of code points that have these two
5031           properties will never change, so something that is not quoted in
5032           v5.16 will never need to be quoted in any future Perl release.
5033           (Not all the code points that match Pattern_Syntax have actually
5034           had characters assigned to them; so there is room to grow, but they
5035           are quoted whether assigned or not.  Perl, of course, would never
5036           use an unassigned code point as an actual metacharacter.)
5037
5038           Quoting characters that have the other 3 properties is done to
5039           enhance the readability of the regular expression and not because
5040           they actually need to be quoted for regular expression purposes
5041           (characters with the White_Space property are likely to be
5042           indistinguishable on the page or screen from those with the
5043           Pattern_White_Space property; and the other two properties contain
5044           non-printing characters).
5045
5046       rand EXPR
5047       rand
5048           Returns a random fractional number greater than or equal to 0 and
5049           less than the value of EXPR.  (EXPR should be positive.)  If EXPR
5050           is omitted, the value 1 is used.  Currently EXPR with the value 0
5051           is also special-cased as 1 (this was undocumented before Perl 5.8.0
5052           and is subject to change in future versions of Perl).
5053           Automatically calls "srand" unless "srand" has already been called.
5054           See also "srand".
5055
5056           Apply "int" to the value returned by "rand" if you want random
5057           integers instead of random fractional numbers.  For example,
5058
5059               int(rand(10))
5060
5061           returns a random integer between 0 and 9, inclusive.
5062
5063           (Note: If your rand function consistently returns numbers that are
5064           too large or too small, then your version of Perl was probably
5065           compiled with the wrong number of RANDBITS.)
5066
5067           "rand" is not cryptographically secure.  You should not rely on it
5068           in security-sensitive situations.  As of this writing, a number of
5069           third-party CPAN modules offer random number generators intended by
5070           their authors to be cryptographically secure, including:
5071           Data::Entropy, Crypt::Random, Math::Random::Secure, and
5072           Math::TrulyRandom.
5073
5074       read FILEHANDLE,SCALAR,LENGTH,OFFSET
5075       read FILEHANDLE,SCALAR,LENGTH
5076           Attempts to read LENGTH characters of data into variable SCALAR
5077           from the specified FILEHANDLE.  Returns the number of characters
5078           actually read, 0 at end of file, or undef if there was an error (in
5079           the latter case $! is also set).  SCALAR will be grown or shrunk so
5080           that the last character actually read is the last character of the
5081           scalar after the read.
5082
5083           An OFFSET may be specified to place the read data at some place in
5084           the string other than the beginning.  A negative OFFSET specifies
5085           placement at that many characters counting backwards from the end
5086           of the string.  A positive OFFSET greater than the length of SCALAR
5087           results in the string being padded to the required size with "\0"
5088           bytes before the result of the read is appended.
5089
5090           The call is implemented in terms of either Perl's or your system's
5091           native fread(3) library function, via the PerlIO layers applied to
5092           the handle.  To get a true read(2) system call, see sysread.
5093
5094           Note the characters: depending on the status of the filehandle,
5095           either (8-bit) bytes or characters are read.  By default, all
5096           filehandles operate on bytes, but for example if the filehandle has
5097           been opened with the ":utf8" I/O layer (see "open", and the open
5098           pragma), the I/O will operate on UTF8-encoded Unicode characters,
5099           not bytes.  Similarly for the ":encoding" layer: in that case
5100           pretty much any characters can be read.
5101
5102       readdir DIRHANDLE
5103           Returns the next directory entry for a directory opened by
5104           "opendir".  If used in list context, returns all the rest of the
5105           entries in the directory.  If there are no more entries, returns
5106           the undefined value in scalar context and the empty list in list
5107           context.
5108
5109           If you're planning to filetest the return values out of a
5110           "readdir", you'd better prepend the directory in question.
5111           Otherwise, because we didn't "chdir" there, it would have been
5112           testing the wrong file.
5113
5114               opendir(my $dh, $some_dir) || die "Can't opendir $some_dir: $!";
5115               my @dots = grep { /^\./ && -f "$some_dir/$_" } readdir($dh);
5116               closedir $dh;
5117
5118           As of Perl 5.12 you can use a bare "readdir" in a "while" loop,
5119           which will set $_ on every iteration.  If either a "readdir"
5120           expression or an explicit assignment of a "readdir" expression to a
5121           scalar is used as a "while"/"for" condition, then the condition
5122           actually tests for definedness of the expression's value, not for
5123           its regular truth value.
5124
5125               opendir(my $dh, $some_dir) || die "Can't open $some_dir: $!";
5126               while (readdir $dh) {
5127                   print "$some_dir/$_\n";
5128               }
5129               closedir $dh;
5130
5131           To avoid confusing would-be users of your code who are running
5132           earlier versions of Perl with mysterious failures, put this sort of
5133           thing at the top of your file to signal that your code will work
5134           only on Perls of a recent vintage:
5135
5136               use 5.012; # so readdir assigns to $_ in a lone while test
5137
5138       readline EXPR
5139       readline
5140           Reads from the filehandle whose typeglob is contained in EXPR (or
5141           from *ARGV if EXPR is not provided).  In scalar context, each call
5142           reads and returns the next line until end-of-file is reached,
5143           whereupon the subsequent call returns "undef".  In list context,
5144           reads until end-of-file is reached and returns a list of lines.
5145           Note that the notion of "line" used here is whatever you may have
5146           defined with $/ (or $INPUT_RECORD_SEPARATOR in English).  See "$/"
5147           in perlvar.
5148
5149           When $/ is set to "undef", when "readline" is in scalar context
5150           (i.e., file slurp mode), and when an empty file is read, it returns
5151           '' the first time, followed by "undef" subsequently.
5152
5153           This is the internal function implementing the "<EXPR>" operator,
5154           but you can use it directly.  The "<EXPR>" operator is discussed in
5155           more detail in "I/O Operators" in perlop.
5156
5157               my $line = <STDIN>;
5158               my $line = readline(STDIN);    # same thing
5159
5160           If "readline" encounters an operating system error, $! will be set
5161           with the corresponding error message.  It can be helpful to check
5162           $! when you are reading from filehandles you don't trust, such as a
5163           tty or a socket.  The following example uses the operator form of
5164           "readline" and dies if the result is not defined.
5165
5166               while ( ! eof($fh) ) {
5167                   defined( $_ = readline $fh ) or die "readline failed: $!";
5168                   ...
5169               }
5170
5171           Note that you have can't handle "readline" errors that way with the
5172           "ARGV" filehandle.  In that case, you have to open each element of
5173           @ARGV yourself since "eof" handles "ARGV" differently.
5174
5175               foreach my $arg (@ARGV) {
5176                   open(my $fh, $arg) or warn "Can't open $arg: $!";
5177
5178                   while ( ! eof($fh) ) {
5179                       defined( $_ = readline $fh )
5180                           or die "readline failed for $arg: $!";
5181                       ...
5182                   }
5183               }
5184
5185           Like the "<EXPR>" operator, if a "readline" expression is used as
5186           the condition of a "while" or "for" loop, then it will be
5187           implicitly assigned to $_.  If either a "readline" expression or an
5188           explicit assignment of a "readline" expression to a scalar is used
5189           as a "while"/"for" condition, then the condition actually tests for
5190           definedness of the expression's value, not for its regular truth
5191           value.
5192
5193       readlink EXPR
5194       readlink
5195           Returns the value of a symbolic link, if symbolic links are
5196           implemented.  If not, raises an exception.  If there is a system
5197           error, returns the undefined value and sets $! (errno).  If EXPR is
5198           omitted, uses $_.
5199
5200           Portability issues: "readlink" in perlport.
5201
5202       readpipe EXPR
5203       readpipe
5204           EXPR is executed as a system command.  The collected standard
5205           output of the command is returned.  In scalar context, it comes
5206           back as a single (potentially multi-line) string.  In list context,
5207           returns a list of lines (however you've defined lines with $/ (or
5208           $INPUT_RECORD_SEPARATOR in English)).  This is the internal
5209           function implementing the "qx/EXPR/" operator, but you can use it
5210           directly.  The "qx/EXPR/" operator is discussed in more detail in
5211           ""qx/STRING/"" in perlop.  If EXPR is omitted, uses $_.
5212
5213       recv SOCKET,SCALAR,LENGTH,FLAGS
5214           Receives a message on a socket.  Attempts to receive LENGTH
5215           characters of data into variable SCALAR from the specified SOCKET
5216           filehandle.  SCALAR will be grown or shrunk to the length actually
5217           read.  Takes the same flags as the system call of the same name.
5218           Returns the address of the sender if SOCKET's protocol supports
5219           this; returns an empty string otherwise.  If there's an error,
5220           returns the undefined value.  This call is actually implemented in
5221           terms of the recvfrom(2) system call.  See "UDP: Message Passing"
5222           in perlipc for examples.
5223
5224           Note that if the socket has been marked as ":utf8", "recv" will
5225           throw an exception.  The ":encoding(...)" layer implicitly
5226           introduces the ":utf8" layer.  See "binmode".
5227
5228       redo LABEL
5229       redo EXPR
5230       redo
5231           The "redo" command restarts the loop block without evaluating the
5232           conditional again.  The "continue" block, if any, is not executed.
5233           If the LABEL is omitted, the command refers to the innermost
5234           enclosing loop.  The "redo EXPR" form, available starting in Perl
5235           5.18.0, allows a label name to be computed at run time, and is
5236           otherwise identical to "redo LABEL".  Programs that want to lie to
5237           themselves about what was just input normally use this command:
5238
5239               # a simpleminded Pascal comment stripper
5240               # (warning: assumes no { or } in strings)
5241               LINE: while (<STDIN>) {
5242                   while (s|({.*}.*){.*}|$1 |) {}
5243                   s|{.*}| |;
5244                   if (s|{.*| |) {
5245                       my $front = $_;
5246                       while (<STDIN>) {
5247                           if (/}/) {  # end of comment?
5248                               s|^|$front\{|;
5249                               redo LINE;
5250                           }
5251                       }
5252                   }
5253                   print;
5254               }
5255
5256           "redo" cannot return a value from a block that typically returns a
5257           value, such as "eval {}", "sub {}", or "do {}". It will perform its
5258           flow control behavior, which precludes any return value. It should
5259           not be used to exit a "grep" or "map" operation.
5260
5261           Note that a block by itself is semantically identical to a loop
5262           that executes once.  Thus "redo" inside such a block will
5263           effectively turn it into a looping construct.
5264
5265           See also "continue" for an illustration of how "last", "next", and
5266           "redo" work.
5267
5268           Unlike most named operators, this has the same precedence as
5269           assignment.  It is also exempt from the looks-like-a-function rule,
5270           so "redo ("foo")."bar"" will cause "bar" to be part of the argument
5271           to "redo".
5272
5273       ref EXPR
5274       ref Examines the value of EXPR, expecting it to be a reference, and
5275           returns a string giving information about the reference and the
5276           type of referent.  If EXPR is not specified, $_ will be used.
5277
5278           If the operand is not a reference, then the empty string will be
5279           returned.  An empty string will only be returned in this situation.
5280           "ref" is often useful to just test whether a value is a reference,
5281           which can be done by comparing the result to the empty string.  It
5282           is a common mistake to use the result of "ref" directly as a truth
5283           value: this goes wrong because 0 (which is false) can be returned
5284           for a reference.
5285
5286           If the operand is a reference to a blessed object, then the name of
5287           the class into which the referent is blessed will be returned.
5288           "ref" doesn't care what the physical type of the referent is;
5289           blessing takes precedence over such concerns.  Beware that exact
5290           comparison of "ref" results against a class name doesn't perform a
5291           class membership test: a class's members also include objects
5292           blessed into subclasses, for which "ref" will return the name of
5293           the subclass.  Also beware that class names can clash with the
5294           built-in type names (described below).
5295
5296           If the operand is a reference to an unblessed object, then the
5297           return value indicates the type of object.  If the unblessed
5298           referent is not a scalar, then the return value will be one of the
5299           strings "ARRAY", "HASH", "CODE", "FORMAT", or "IO", indicating only
5300           which kind of object it is.  If the unblessed referent is a scalar,
5301           then the return value will be one of the strings "SCALAR",
5302           "VSTRING", "REF", "GLOB", "LVALUE", or "REGEXP", depending on the
5303           kind of value the scalar currently has.   But note that "qr//"
5304           scalars are created already blessed, so "ref qr/.../" will likely
5305           return "Regexp".  Beware that these built-in type names can also be
5306           used as class names, so "ref" returning one of these names doesn't
5307           unambiguously indicate that the referent is of the kind to which
5308           the name refers.
5309
5310           The ambiguity between built-in type names and class names
5311           significantly limits the utility of "ref".  For unambiguous
5312           information, use "Scalar::Util::blessed()" for information about
5313           blessing, and "Scalar::Util::reftype()" for information about
5314           physical types.  Use the "isa" method for class membership tests,
5315           though one must be sure of blessedness before attempting a method
5316           call.
5317
5318           See also perlref and perlobj.
5319
5320       rename OLDNAME,NEWNAME
5321           Changes the name of a file; an existing file NEWNAME will be
5322           clobbered.  Returns true for success, false otherwise.
5323
5324           Behavior of this function varies wildly depending on your system
5325           implementation.  For example, it will usually not work across file
5326           system boundaries, even though the system mv command sometimes
5327           compensates for this.  Other restrictions include whether it works
5328           on directories, open files, or pre-existing files.  Check perlport
5329           and either the rename(2) manpage or equivalent system documentation
5330           for details.
5331
5332           For a platform independent "move" function look at the File::Copy
5333           module.
5334
5335           Portability issues: "rename" in perlport.
5336
5337       require VERSION
5338       require EXPR
5339       require
5340           Demands a version of Perl specified by VERSION, or demands some
5341           semantics specified by EXPR or by $_ if EXPR is not supplied.
5342
5343           VERSION may be either a literal such as v5.24.1, which will be
5344           compared to $^V (or $PERL_VERSION in English), or a numeric
5345           argument of the form 5.024001, which will be compared to $]. An
5346           exception is raised if VERSION is greater than the version of the
5347           current Perl interpreter.  Compare with "use", which can do a
5348           similar check at compile time.
5349
5350           Specifying VERSION as a numeric argument of the form 5.024001
5351           should generally be avoided as older less readable syntax compared
5352           to v5.24.1. Before perl 5.8.0 (released in 2002), the more verbose
5353           numeric form was the only supported syntax, which is why you might
5354           see it in older code.
5355
5356               require v5.24.1;    # run time version check
5357               require 5.24.1;     # ditto
5358               require 5.024_001;  # ditto; older syntax compatible
5359                                     with perl 5.6
5360
5361           Otherwise, "require" demands that a library file be included if it
5362           hasn't already been included.  The file is included via the do-FILE
5363           mechanism, which is essentially just a variety of "eval" with the
5364           caveat that lexical variables in the invoking script will be
5365           invisible to the included code.  If it were implemented in pure
5366           Perl, it would have semantics similar to the following:
5367
5368               use Carp 'croak';
5369               use version;
5370
5371               sub require {
5372                   my ($filename) = @_;
5373                   if ( my $version = eval { version->parse($filename) } ) {
5374                       if ( $version > $^V ) {
5375                          my $vn = $version->normal;
5376                          croak "Perl $vn required--this is only $^V, stopped";
5377                       }
5378                       return 1;
5379                   }
5380
5381                   if (exists $INC{$filename}) {
5382                       return 1 if $INC{$filename};
5383                       croak "Compilation failed in require";
5384                   }
5385
5386                   foreach $prefix (@INC) {
5387                       if (ref($prefix)) {
5388                           #... do other stuff - see text below ....
5389                       }
5390                       # (see text below about possible appending of .pmc
5391                       # suffix to $filename)
5392                       my $realfilename = "$prefix/$filename";
5393                       next if ! -e $realfilename || -d _ || -b _;
5394                       $INC{$filename} = $realfilename;
5395                       my $result = do($realfilename);
5396                                    # but run in caller's namespace
5397
5398                       if (!defined $result) {
5399                           $INC{$filename} = undef;
5400                           croak $@ ? "$@Compilation failed in require"
5401                                    : "Can't locate $filename: $!\n";
5402                       }
5403                       if (!$result) {
5404                           delete $INC{$filename};
5405                           croak "$filename did not return true value";
5406                       }
5407                       $! = 0;
5408                       return $result;
5409                   }
5410                   croak "Can't locate $filename in \@INC ...";
5411               }
5412
5413           Note that the file will not be included twice under the same
5414           specified name.
5415
5416           The file must return true as the last statement to indicate
5417           successful execution of any initialization code, so it's customary
5418           to end such a file with "1;" unless you're sure it'll return true
5419           otherwise.  But it's better just to put the "1;", in case you add
5420           more statements.
5421
5422           If EXPR is a bareword, "require" assumes a .pm extension and
5423           replaces "::" with "/" in the filename for you, to make it easy to
5424           load standard modules.  This form of loading of modules does not
5425           risk altering your namespace, however it will autovivify the stash
5426           for the required module.
5427
5428           In other words, if you try this:
5429
5430                   require Foo::Bar;     # a splendid bareword
5431
5432           The require function will actually look for the Foo/Bar.pm file in
5433           the directories specified in the @INC array, and it will autovivify
5434           the "Foo::Bar::" stash at compile time.
5435
5436           But if you try this:
5437
5438                   my $class = 'Foo::Bar';
5439                   require $class;       # $class is not a bareword
5440               #or
5441                   require "Foo::Bar";   # not a bareword because of the ""
5442
5443           The require function will look for the Foo::Bar file in the @INC
5444           array and will complain about not finding Foo::Bar there.  In this
5445           case you can do:
5446
5447                   eval "require $class";
5448
5449           or you could do
5450
5451                   require "Foo/Bar.pm";
5452
5453           Neither of these forms will autovivify any stashes at compile time
5454           and only have run time effects.
5455
5456           Now that you understand how "require" looks for files with a
5457           bareword argument, there is a little extra functionality going on
5458           behind the scenes.  Before "require" looks for a .pm extension, it
5459           will first look for a similar filename with a .pmc extension.  If
5460           this file is found, it will be loaded in place of any file ending
5461           in a .pm extension. This applies to both the explicit "require
5462           "Foo/Bar.pm";" form and the "require Foo::Bar;" form.
5463
5464           You can also insert hooks into the import facility by putting Perl
5465           code directly into the @INC array.  There are three forms of hooks:
5466           subroutine references, array references, and blessed objects.
5467
5468           Subroutine references are the simplest case.  When the inclusion
5469           system walks through @INC and encounters a subroutine, this
5470           subroutine gets called with two parameters, the first a reference
5471           to itself, and the second the name of the file to be included
5472           (e.g., Foo/Bar.pm).  The subroutine should return either nothing or
5473           else a list of up to four values in the following order:
5474
5475           1.  A reference to a scalar, containing any initial source code to
5476               prepend to the file or generator output.
5477
5478           2.  A filehandle, from which the file will be read.
5479
5480           3.  A reference to a subroutine.  If there is no filehandle
5481               (previous item), then this subroutine is expected to generate
5482               one line of source code per call, writing the line into $_ and
5483               returning 1, then finally at end of file returning 0.  If there
5484               is a filehandle, then the subroutine will be called to act as a
5485               simple source filter, with the line as read in $_.  Again,
5486               return 1 for each valid line, and 0 after all lines have been
5487               returned.  For historical reasons the subroutine will receive a
5488               meaningless argument (in fact always the numeric value zero) as
5489               $_[0].
5490
5491           4.  Optional state for the subroutine.  The state is passed in as
5492               $_[1].
5493
5494           If an empty list, "undef", or nothing that matches the first 3
5495           values above is returned, then "require" looks at the remaining
5496           elements of @INC.  Note that this filehandle must be a real
5497           filehandle (strictly a typeglob or reference to a typeglob, whether
5498           blessed or unblessed); tied filehandles will be ignored and
5499           processing will stop there.
5500
5501           If the hook is an array reference, its first element must be a
5502           subroutine reference.  This subroutine is called as above, but the
5503           first parameter is the array reference.  This lets you indirectly
5504           pass arguments to the subroutine.
5505
5506           In other words, you can write:
5507
5508               push @INC, \&my_sub;
5509               sub my_sub {
5510                   my ($coderef, $filename) = @_;  # $coderef is \&my_sub
5511                   ...
5512               }
5513
5514           or:
5515
5516               push @INC, [ \&my_sub, $x, $y, ... ];
5517               sub my_sub {
5518                   my ($arrayref, $filename) = @_;
5519                   # Retrieve $x, $y, ...
5520                   my (undef, @parameters) = @$arrayref;
5521                   ...
5522               }
5523
5524           If the hook is an object, it must provide an "INC" method that will
5525           be called as above, the first parameter being the object itself.
5526           (Note that you must fully qualify the sub's name, as unqualified
5527           "INC" is always forced into package "main".)  Here is a typical
5528           code layout:
5529
5530               # In Foo.pm
5531               package Foo;
5532               sub new { ... }
5533               sub Foo::INC {
5534                   my ($self, $filename) = @_;
5535                   ...
5536               }
5537
5538               # In the main program
5539               push @INC, Foo->new(...);
5540
5541           These hooks are also permitted to set the %INC entry corresponding
5542           to the files they have loaded.  See "%INC" in perlvar.
5543
5544           For a yet-more-powerful import facility, see "use" and perlmod.
5545
5546       reset EXPR
5547       reset
5548           Generally used in a "continue" block at the end of a loop to clear
5549           variables and reset "m?pattern?" searches so that they work again.
5550           The expression is interpreted as a list of single characters
5551           (hyphens allowed for ranges).  All variables (scalars, arrays, and
5552           hashes) in the current package beginning with one of those letters
5553           are reset to their pristine state.  If the expression is omitted,
5554           one-match searches ("m?pattern?") are reset to match again.  Only
5555           resets variables or searches in the current package.  Always
5556           returns 1.  Examples:
5557
5558               reset 'X';      # reset all X variables
5559               reset 'a-z';    # reset lower case variables
5560               reset;          # just reset m?one-time? searches
5561
5562           Resetting "A-Z" is not recommended because you'll wipe out your
5563           @ARGV and @INC arrays and your %ENV hash.
5564
5565           Resets only package variables; lexical variables are unaffected,
5566           but they clean themselves up on scope exit anyway, so you'll
5567           probably want to use them instead.  See "my".
5568
5569       return EXPR
5570       return
5571           Returns from a subroutine, "eval", "do FILE", "sort" block or regex
5572           eval block (but not a "grep", "map", or "do BLOCK" block) with the
5573           value given in EXPR.  Evaluation of EXPR may be in list, scalar, or
5574           void context, depending on how the return value will be used, and
5575           the context may vary from one execution to the next (see
5576           "wantarray").  If no EXPR is given, returns an empty list in list
5577           context, the undefined value in scalar context, and (of course)
5578           nothing at all in void context.
5579
5580           (In the absence of an explicit "return", a subroutine, "eval", or
5581           "do FILE" automatically returns the value of the last expression
5582           evaluated.)
5583
5584           Unlike most named operators, this is also exempt from the looks-
5585           like-a-function rule, so "return ("foo")."bar"" will cause "bar" to
5586           be part of the argument to "return".
5587
5588       reverse LIST
5589           In list context, returns a list value consisting of the elements of
5590           LIST in the opposite order.  In scalar context, concatenates the
5591           elements of LIST and returns a string value with all characters in
5592           the opposite order.
5593
5594               print join(", ", reverse "world", "Hello"); # Hello, world
5595
5596               print scalar reverse "dlrow ,", "olleH";    # Hello, world
5597
5598           Used without arguments in scalar context, "reverse" reverses $_.
5599
5600               $_ = "dlrow ,olleH";
5601               print reverse;                         # No output, list context
5602               print scalar reverse;                  # Hello, world
5603
5604           Note that reversing an array to itself (as in "@a = reverse @a")
5605           will preserve non-existent elements whenever possible; i.e., for
5606           non-magical arrays or for tied arrays with "EXISTS" and "DELETE"
5607           methods.
5608
5609           This operator is also handy for inverting a hash, although there
5610           are some caveats.  If a value is duplicated in the original hash,
5611           only one of those can be represented as a key in the inverted hash.
5612           Also, this has to unwind one hash and build a whole new one, which
5613           may take some time on a large hash, such as from a DBM file.
5614
5615               my %by_name = reverse %by_address;  # Invert the hash
5616
5617       rewinddir DIRHANDLE
5618           Sets the current position to the beginning of the directory for the
5619           "readdir" routine on DIRHANDLE.
5620
5621           Portability issues: "rewinddir" in perlport.
5622
5623       rindex STR,SUBSTR,POSITION
5624       rindex STR,SUBSTR
5625           Works just like "index" except that it returns the position of the
5626           last occurrence of SUBSTR in STR.  If POSITION is specified,
5627           returns the last occurrence beginning at or before that position.
5628
5629       rmdir FILENAME
5630       rmdir
5631           Deletes the directory specified by FILENAME if that directory is
5632           empty.  If it succeeds it returns true; otherwise it returns false
5633           and sets $! (errno).  If FILENAME is omitted, uses $_.
5634
5635           To remove a directory tree recursively ("rm -rf" on Unix) look at
5636           the "rmtree" function of the File::Path module.
5637
5638       s///
5639           The substitution operator.  See "Regexp Quote-Like Operators" in
5640           perlop.
5641
5642       say FILEHANDLE LIST
5643       say FILEHANDLE
5644       say LIST
5645       say Just like "print", but implicitly appends a newline at the end of
5646           the LIST instead of any value "$\" might have.  To use FILEHANDLE
5647           without a LIST to print the contents of $_ to it, you must use a
5648           bareword filehandle like "FH", not an indirect one like $fh.
5649
5650           "say" is available only if the "say" feature is enabled or if it is
5651           prefixed with "CORE::".  The "say" feature is enabled automatically
5652           with a "use v5.10" (or higher) declaration in the current scope.
5653
5654       scalar EXPR
5655           Forces EXPR to be interpreted in scalar context and returns the
5656           value of EXPR.
5657
5658               my @counts = ( scalar @a, scalar @b, scalar @c );
5659
5660           There is no equivalent operator to force an expression to be
5661           interpolated in list context because in practice, this is never
5662           needed.  If you really wanted to do so, however, you could use the
5663           construction "@{[ (some expression) ]}", but usually a simple
5664           "(some expression)" suffices.
5665
5666           Because "scalar" is a unary operator, if you accidentally use a
5667           parenthesized list for the EXPR, this behaves as a scalar comma
5668           expression, evaluating all but the last element in void context and
5669           returning the final element evaluated in scalar context.  This is
5670           seldom what you want.
5671
5672           The following single statement:
5673
5674               print uc(scalar(foo(), $bar)), $baz;
5675
5676           is the moral equivalent of these two:
5677
5678               foo();
5679               print(uc($bar), $baz);
5680
5681           See perlop for more details on unary operators and the comma
5682           operator, and perldata for details on evaluating a hash in scalar
5683           contex.
5684
5685       seek FILEHANDLE,POSITION,WHENCE
5686           Sets FILEHANDLE's position, just like the fseek(3) call of C
5687           "stdio".  FILEHANDLE may be an expression whose value gives the
5688           name of the filehandle.  The values for WHENCE are 0 to set the new
5689           position in bytes to POSITION; 1 to set it to the current position
5690           plus POSITION; and 2 to set it to EOF plus POSITION, typically
5691           negative.  For WHENCE you may use the constants "SEEK_SET",
5692           "SEEK_CUR", and "SEEK_END" (start of the file, current position,
5693           end of the file) from the Fcntl module.  Returns 1 on success,
5694           false otherwise.
5695
5696           Note the emphasis on bytes: even if the filehandle has been set to
5697           operate on characters (for example using the ":encoding(UTF-8)" I/O
5698           layer), the "seek", "tell", and "sysseek" family of functions use
5699           byte offsets, not character offsets, because seeking to a character
5700           offset would be very slow in a UTF-8 file.
5701
5702           If you want to position the file for "sysread" or "syswrite", don't
5703           use "seek", because buffering makes its effect on the file's read-
5704           write position unpredictable and non-portable.  Use "sysseek"
5705           instead.
5706
5707           Due to the rules and rigors of ANSI C, on some systems you have to
5708           do a seek whenever you switch between reading and writing.  Amongst
5709           other things, this may have the effect of calling stdio's
5710           clearerr(3).  A WHENCE of 1 ("SEEK_CUR") is useful for not moving
5711           the file position:
5712
5713               seek($fh, 0, 1);
5714
5715           This is also useful for applications emulating "tail -f".  Once you
5716           hit EOF on your read and then sleep for a while, you (probably)
5717           have to stick in a dummy "seek" to reset things.  The "seek"
5718           doesn't change the position, but it does clear the end-of-file
5719           condition on the handle, so that the next "readline FILE" makes
5720           Perl try again to read something.  (We hope.)
5721
5722           If that doesn't work (some I/O implementations are particularly
5723           cantankerous), you might need something like this:
5724
5725               for (;;) {
5726                   for ($curpos = tell($fh); $_ = readline($fh);
5727                        $curpos = tell($fh)) {
5728                       # search for some stuff and put it into files
5729                   }
5730                   sleep($for_a_while);
5731                   seek($fh, $curpos, 0);
5732               }
5733
5734       seekdir DIRHANDLE,POS
5735           Sets the current position for the "readdir" routine on DIRHANDLE.
5736           POS must be a value returned by "telldir".  "seekdir" also has the
5737           same caveats about possible directory compaction as the
5738           corresponding system library routine.
5739
5740       select FILEHANDLE
5741       select
5742           Returns the currently selected filehandle.  If FILEHANDLE is
5743           supplied, sets the new current default filehandle for output.  This
5744           has two effects: first, a "write" or a "print" without a filehandle
5745           default to this FILEHANDLE.  Second, references to variables
5746           related to output will refer to this output channel.
5747
5748           For example, to set the top-of-form format for more than one output
5749           channel, you might do the following:
5750
5751               select(REPORT1);
5752               $^ = 'report1_top';
5753               select(REPORT2);
5754               $^ = 'report2_top';
5755
5756           FILEHANDLE may be an expression whose value gives the name of the
5757           actual filehandle.  Thus:
5758
5759               my $oldfh = select(STDERR); $| = 1; select($oldfh);
5760
5761           Some programmers may prefer to think of filehandles as objects with
5762           methods, preferring to write the last example as:
5763
5764               STDERR->autoflush(1);
5765
5766           (Prior to Perl version 5.14, you have to "use IO::Handle;"
5767           explicitly first.)
5768
5769           Portability issues: "select" in perlport.
5770
5771       select RBITS,WBITS,EBITS,TIMEOUT
5772           This calls the select(2) syscall with the bit masks specified,
5773           which can be constructed using "fileno" and "vec", along these
5774           lines:
5775
5776               my $rin = my $win = my $ein = '';
5777               vec($rin, fileno(STDIN),  1) = 1;
5778               vec($win, fileno(STDOUT), 1) = 1;
5779               $ein = $rin | $win;
5780
5781           If you want to select on many filehandles, you may wish to write a
5782           subroutine like this:
5783
5784               sub fhbits {
5785                   my @fhlist = @_;
5786                   my $bits = "";
5787                   for my $fh (@fhlist) {
5788                       vec($bits, fileno($fh), 1) = 1;
5789                   }
5790                   return $bits;
5791               }
5792               my $rin = fhbits(\*STDIN, $tty, $mysock);
5793
5794           The usual idiom is:
5795
5796            my ($nfound, $timeleft) =
5797              select(my $rout = $rin, my $wout = $win, my $eout = $ein,
5798                                                                     $timeout);
5799
5800           or to block until something becomes ready just do this
5801
5802            my $nfound =
5803              select(my $rout = $rin, my $wout = $win, my $eout = $ein, undef);
5804
5805           Most systems do not bother to return anything useful in $timeleft,
5806           so calling "select" in scalar context just returns $nfound.
5807
5808           Any of the bit masks can also be "undef".  The timeout, if
5809           specified, is in seconds, which may be fractional.  Note: not all
5810           implementations are capable of returning the $timeleft.  If not,
5811           they always return $timeleft equal to the supplied $timeout.
5812
5813           You can effect a sleep of 250 milliseconds this way:
5814
5815               select(undef, undef, undef, 0.25);
5816
5817           Note that whether "select" gets restarted after signals (say,
5818           SIGALRM) is implementation-dependent.  See also perlport for notes
5819           on the portability of "select".
5820
5821           On error, "select" behaves just like select(2): it returns "-1" and
5822           sets $!.
5823
5824           On some Unixes, select(2) may report a socket file descriptor as
5825           "ready for reading" even when no data is available, and thus any
5826           subsequent "read" would block.  This can be avoided if you always
5827           use "O_NONBLOCK" on the socket.  See select(2) and fcntl(2) for
5828           further details.
5829
5830           The standard "IO::Select" module provides a user-friendlier
5831           interface to "select", mostly because it does all the bit-mask work
5832           for you.
5833
5834           WARNING: One should not attempt to mix buffered I/O (like "read" or
5835           "readline") with "select", except as permitted by POSIX, and even
5836           then only on POSIX systems.  You have to use "sysread" instead.
5837
5838           Portability issues: "select" in perlport.
5839
5840       semctl ID,SEMNUM,CMD,ARG
5841           Calls the System V IPC function semctl(2).  You'll probably have to
5842           say
5843
5844               use IPC::SysV;
5845
5846           first to get the correct constant definitions.  If CMD is IPC_STAT
5847           or GETALL, then ARG must be a variable that will hold the returned
5848           semid_ds structure or semaphore value array.  Returns like "ioctl":
5849           the undefined value for error, ""0 but true"" for zero, or the
5850           actual return value otherwise.  The ARG must consist of a vector of
5851           native short integers, which may be created with
5852           "pack("s!",(0)x$nsem)".  See also "SysV IPC" in perlipc and the
5853           documentation for "IPC::SysV" and "IPC::Semaphore".
5854
5855           Portability issues: "semctl" in perlport.
5856
5857       semget KEY,NSEMS,FLAGS
5858           Calls the System V IPC function semget(2).  Returns the semaphore
5859           id, or the undefined value on error.  See also "SysV IPC" in
5860           perlipc and the documentation for "IPC::SysV" and "IPC::Semaphore".
5861
5862           Portability issues: "semget" in perlport.
5863
5864       semop KEY,OPSTRING
5865           Calls the System V IPC function semop(2) for semaphore operations
5866           such as signalling and waiting.  OPSTRING must be a packed array of
5867           semop structures.  Each semop structure can be generated with
5868           "pack("s!3", $semnum, $semop, $semflag)".  The length of OPSTRING
5869           implies the number of semaphore operations.  Returns true if
5870           successful, false on error.  As an example, the following code
5871           waits on semaphore $semnum of semaphore id $semid:
5872
5873               my $semop = pack("s!3", $semnum, -1, 0);
5874               die "Semaphore trouble: $!\n" unless semop($semid, $semop);
5875
5876           To signal the semaphore, replace "-1" with 1.  See also "SysV IPC"
5877           in perlipc and the documentation for "IPC::SysV" and
5878           "IPC::Semaphore".
5879
5880           Portability issues: "semop" in perlport.
5881
5882       send SOCKET,MSG,FLAGS,TO
5883       send SOCKET,MSG,FLAGS
5884           Sends a message on a socket.  Attempts to send the scalar MSG to
5885           the SOCKET filehandle.  Takes the same flags as the system call of
5886           the same name.  On unconnected sockets, you must specify a
5887           destination to send to, in which case it does a sendto(2) syscall.
5888           Returns the number of characters sent, or the undefined value on
5889           error.  The sendmsg(2) syscall is currently unimplemented.  See
5890           "UDP: Message Passing" in perlipc for examples.
5891
5892           Note that if the socket has been marked as ":utf8", "send" will
5893           throw an exception.  The ":encoding(...)" layer implicitly
5894           introduces the ":utf8" layer.  See "binmode".
5895
5896       setpgrp PID,PGRP
5897           Sets the current process group for the specified PID, 0 for the
5898           current process.  Raises an exception when used on a machine that
5899           doesn't implement POSIX setpgid(2) or BSD setpgrp(2).  If the
5900           arguments are omitted, it defaults to "0,0".  Note that the BSD 4.2
5901           version of "setpgrp" does not accept any arguments, so only
5902           "setpgrp(0,0)" is portable.  See also "POSIX::setsid()".
5903
5904           Portability issues: "setpgrp" in perlport.
5905
5906       setpriority WHICH,WHO,PRIORITY
5907           Sets the current priority for a process, a process group, or a
5908           user.  (See setpriority(2).)  Raises an exception when used on a
5909           machine that doesn't implement setpriority(2).
5910
5911           "WHICH" can be any of "PRIO_PROCESS", "PRIO_PGRP" or "PRIO_USER"
5912           imported from "RESOURCE CONSTANTS" in POSIX.
5913
5914           Portability issues: "setpriority" in perlport.
5915
5916       setsockopt SOCKET,LEVEL,OPTNAME,OPTVAL
5917           Sets the socket option requested.  Returns "undef" on error.  Use
5918           integer constants provided by the "Socket" module for LEVEL and
5919           OPNAME.  Values for LEVEL can also be obtained from getprotobyname.
5920           OPTVAL might either be a packed string or an integer.  An integer
5921           OPTVAL is shorthand for pack("i", OPTVAL).
5922
5923           An example disabling Nagle's algorithm on a socket:
5924
5925               use Socket qw(IPPROTO_TCP TCP_NODELAY);
5926               setsockopt($socket, IPPROTO_TCP, TCP_NODELAY, 1);
5927
5928           Portability issues: "setsockopt" in perlport.
5929
5930       shift ARRAY
5931       shift
5932           Shifts the first value of the array off and returns it, shortening
5933           the array by 1 and moving everything down.  If there are no
5934           elements in the array, returns the undefined value.  If ARRAY is
5935           omitted, shifts the @_ array within the lexical scope of
5936           subroutines and formats, and the @ARGV array outside a subroutine
5937           and also within the lexical scopes established by the "eval
5938           STRING", "BEGIN {}", "INIT {}", "CHECK {}", "UNITCHECK {}", and
5939           "END {}" constructs.
5940
5941           Starting with Perl 5.14, an experimental feature allowed "shift" to
5942           take a scalar expression. This experiment has been deemed
5943           unsuccessful, and was removed as of Perl 5.24.
5944
5945           See also "unshift", "push", and "pop".  "shift" and "unshift" do
5946           the same thing to the left end of an array that "pop" and "push" do
5947           to the right end.
5948
5949       shmctl ID,CMD,ARG
5950           Calls the System V IPC function shmctl.  You'll probably have to
5951           say
5952
5953               use IPC::SysV;
5954
5955           first to get the correct constant definitions.  If CMD is
5956           "IPC_STAT", then ARG must be a variable that will hold the returned
5957           "shmid_ds" structure.  Returns like ioctl: "undef" for error; "0
5958           but true" for zero; and the actual return value otherwise.  See
5959           also "SysV IPC" in perlipc and the documentation for "IPC::SysV".
5960
5961           Portability issues: "shmctl" in perlport.
5962
5963       shmget KEY,SIZE,FLAGS
5964           Calls the System V IPC function shmget.  Returns the shared memory
5965           segment id, or "undef" on error.  See also "SysV IPC" in perlipc
5966           and the documentation for "IPC::SysV".
5967
5968           Portability issues: "shmget" in perlport.
5969
5970       shmread ID,VAR,POS,SIZE
5971       shmwrite ID,STRING,POS,SIZE
5972           Reads or writes the System V shared memory segment ID starting at
5973           position POS for size SIZE by attaching to it, copying in/out, and
5974           detaching from it.  When reading, VAR must be a variable that will
5975           hold the data read.  When writing, if STRING is too long, only SIZE
5976           bytes are used; if STRING is too short, nulls are written to fill
5977           out SIZE bytes.  Return true if successful, false on error.
5978           "shmread" taints the variable.  See also "SysV IPC" in perlipc and
5979           the documentation for "IPC::SysV" and the "IPC::Shareable" module
5980           from CPAN.
5981
5982           Portability issues: "shmread" in perlport and "shmwrite" in
5983           perlport.
5984
5985       shutdown SOCKET,HOW
5986           Shuts down a socket connection in the manner indicated by HOW,
5987           which has the same interpretation as in the syscall of the same
5988           name.
5989
5990               shutdown($socket, 0);    # I/we have stopped reading data
5991               shutdown($socket, 1);    # I/we have stopped writing data
5992               shutdown($socket, 2);    # I/we have stopped using this socket
5993
5994           This is useful with sockets when you want to tell the other side
5995           you're done writing but not done reading, or vice versa.  It's also
5996           a more insistent form of close because it also disables the file
5997           descriptor in any forked copies in other processes.
5998
5999           Returns 1 for success; on error, returns "undef" if the first
6000           argument is not a valid filehandle, or returns 0 and sets $! for
6001           any other failure.
6002
6003       sin EXPR
6004       sin Returns the sine of EXPR (expressed in radians).  If EXPR is
6005           omitted, returns sine of $_.
6006
6007           For the inverse sine operation, you may use the "Math::Trig::asin"
6008           function, or use this relation:
6009
6010               sub asin { atan2($_[0], sqrt(1 - $_[0] * $_[0])) }
6011
6012       sleep EXPR
6013       sleep
6014           Causes the script to sleep for (integer) EXPR seconds, or forever
6015           if no argument is given.  Returns the integer number of seconds
6016           actually slept.
6017
6018           May be interrupted if the process receives a signal such as
6019           "SIGALRM".
6020
6021               eval {
6022                   local $SIG{ALRM} = sub { die "Alarm!\n" };
6023                   sleep;
6024               };
6025               die $@ unless $@ eq "Alarm!\n";
6026
6027           You probably cannot mix "alarm" and "sleep" calls, because "sleep"
6028           is often implemented using "alarm".
6029
6030           On some older systems, it may sleep up to a full second less than
6031           what you requested, depending on how it counts seconds.  Most
6032           modern systems always sleep the full amount.  They may appear to
6033           sleep longer than that, however, because your process might not be
6034           scheduled right away in a busy multitasking system.
6035
6036           For delays of finer granularity than one second, the Time::HiRes
6037           module (from CPAN, and starting from Perl 5.8 part of the standard
6038           distribution) provides "usleep".  You may also use Perl's four-
6039           argument version of "select" leaving the first three arguments
6040           undefined, or you might be able to use the "syscall" interface to
6041           access setitimer(2) if your system supports it.  See perlfaq8 for
6042           details.
6043
6044           See also the POSIX module's "pause" function.
6045
6046       socket SOCKET,DOMAIN,TYPE,PROTOCOL
6047           Opens a socket of the specified kind and attaches it to filehandle
6048           SOCKET.  DOMAIN, TYPE, and PROTOCOL are specified the same as for
6049           the syscall of the same name.  You should "use Socket" first to get
6050           the proper definitions imported.  See the examples in "Sockets:
6051           Client/Server Communication" in perlipc.
6052
6053           On systems that support a close-on-exec flag on files, the flag
6054           will be set for the newly opened file descriptor, as determined by
6055           the value of $^F.  See "$^F" in perlvar.
6056
6057       socketpair SOCKET1,SOCKET2,DOMAIN,TYPE,PROTOCOL
6058           Creates an unnamed pair of sockets in the specified domain, of the
6059           specified type.  DOMAIN, TYPE, and PROTOCOL are specified the same
6060           as for the syscall of the same name.  If unimplemented, raises an
6061           exception.  Returns true if successful.
6062
6063           On systems that support a close-on-exec flag on files, the flag
6064           will be set for the newly opened file descriptors, as determined by
6065           the value of $^F.  See "$^F" in perlvar.
6066
6067           Some systems define "pipe" in terms of "socketpair", in which a
6068           call to "pipe($rdr, $wtr)" is essentially:
6069
6070               use Socket;
6071               socketpair(my $rdr, my $wtr, AF_UNIX, SOCK_STREAM, PF_UNSPEC);
6072               shutdown($rdr, 1);        # no more writing for reader
6073               shutdown($wtr, 0);        # no more reading for writer
6074
6075           See perlipc for an example of socketpair use.  Perl 5.8 and later
6076           will emulate socketpair using IP sockets to localhost if your
6077           system implements sockets but not socketpair.
6078
6079           Portability issues: "socketpair" in perlport.
6080
6081       sort SUBNAME LIST
6082       sort BLOCK LIST
6083       sort LIST
6084           In list context, this sorts the LIST and returns the sorted list
6085           value.  In scalar context, the behaviour of "sort" is undefined.
6086
6087           If SUBNAME or BLOCK is omitted, "sort"s in standard string
6088           comparison order.  If SUBNAME is specified, it gives the name of a
6089           subroutine that returns an integer less than, equal to, or greater
6090           than 0, depending on how the elements of the list are to be
6091           ordered.  (The "<=>" and "cmp" operators are extremely useful in
6092           such routines.)  SUBNAME may be a scalar variable name
6093           (unsubscripted), in which case the value provides the name of (or a
6094           reference to) the actual subroutine to use.  In place of a SUBNAME,
6095           you can provide a BLOCK as an anonymous, in-line sort subroutine.
6096
6097           If the subroutine's prototype is "($$)", the elements to be
6098           compared are passed by reference in @_, as for a normal subroutine.
6099           This is slower than unprototyped subroutines, where the elements to
6100           be compared are passed into the subroutine as the package global
6101           variables $a and $b (see example below).
6102
6103           If the subroutine is an XSUB, the elements to be compared are
6104           pushed on to the stack, the way arguments are usually passed to
6105           XSUBs.  $a and $b are not set.
6106
6107           The values to be compared are always passed by reference and should
6108           not be modified.
6109
6110           You also cannot exit out of the sort block or subroutine using any
6111           of the loop control operators described in perlsyn or with "goto".
6112
6113           When "use locale" (but not "use locale ':not_characters'") is in
6114           effect, "sort LIST" sorts LIST according to the current collation
6115           locale.  See perllocale.
6116
6117           "sort" returns aliases into the original list, much as a for loop's
6118           index variable aliases the list elements.  That is, modifying an
6119           element of a list returned by "sort" (for example, in a "foreach",
6120           "map" or "grep") actually modifies the element in the original
6121           list.  This is usually something to be avoided when writing clear
6122           code.
6123
6124           Historically Perl has varied in whether sorting is stable by
6125           default.  If stability matters, it can be controlled explicitly by
6126           using the sort pragma.
6127
6128           Examples:
6129
6130               # sort lexically
6131               my @articles = sort @files;
6132
6133               # same thing, but with explicit sort routine
6134               my @articles = sort {$a cmp $b} @files;
6135
6136               # now case-insensitively
6137               my @articles = sort {fc($a) cmp fc($b)} @files;
6138
6139               # same thing in reversed order
6140               my @articles = sort {$b cmp $a} @files;
6141
6142               # sort numerically ascending
6143               my @articles = sort {$a <=> $b} @files;
6144
6145               # sort numerically descending
6146               my @articles = sort {$b <=> $a} @files;
6147
6148               # this sorts the %age hash by value instead of key
6149               # using an in-line function
6150               my @eldest = sort { $age{$b} <=> $age{$a} } keys %age;
6151
6152               # sort using explicit subroutine name
6153               sub byage {
6154                   $age{$a} <=> $age{$b};  # presuming numeric
6155               }
6156               my @sortedclass = sort byage @class;
6157
6158               sub backwards { $b cmp $a }
6159               my @harry  = qw(dog cat x Cain Abel);
6160               my @george = qw(gone chased yz Punished Axed);
6161               print sort @harry;
6162                   # prints AbelCaincatdogx
6163               print sort backwards @harry;
6164                   # prints xdogcatCainAbel
6165               print sort @george, 'to', @harry;
6166                   # prints AbelAxedCainPunishedcatchaseddoggonetoxyz
6167
6168               # inefficiently sort by descending numeric compare using
6169               # the first integer after the first = sign, or the
6170               # whole record case-insensitively otherwise
6171
6172               my @new = sort {
6173                   ($b =~ /=(\d+)/)[0] <=> ($a =~ /=(\d+)/)[0]
6174                                       ||
6175                               fc($a)  cmp  fc($b)
6176               } @old;
6177
6178               # same thing, but much more efficiently;
6179               # we'll build auxiliary indices instead
6180               # for speed
6181               my (@nums, @caps);
6182               for (@old) {
6183                   push @nums, ( /=(\d+)/ ? $1 : undef );
6184                   push @caps, fc($_);
6185               }
6186
6187               my @new = @old[ sort {
6188                                      $nums[$b] <=> $nums[$a]
6189                                               ||
6190                                      $caps[$a] cmp $caps[$b]
6191                                    } 0..$#old
6192                             ];
6193
6194               # same thing, but without any temps
6195               my @new = map { $_->[0] }
6196                      sort { $b->[1] <=> $a->[1]
6197                                      ||
6198                             $a->[2] cmp $b->[2]
6199                      } map { [$_, /=(\d+)/, fc($_)] } @old;
6200
6201               # using a prototype allows you to use any comparison subroutine
6202               # as a sort subroutine (including other package's subroutines)
6203               package Other;
6204               sub backwards ($$) { $_[1] cmp $_[0]; }  # $a and $b are
6205                                                        # not set here
6206               package main;
6207               my @new = sort Other::backwards @old;
6208
6209               # guarantee stability
6210               use sort 'stable';
6211               my @new = sort { substr($a, 3, 5) cmp substr($b, 3, 5) } @old;
6212
6213           Warning: syntactical care is required when sorting the list
6214           returned from a function.  If you want to sort the list returned by
6215           the function call "find_records(@key)", you can use:
6216
6217               my @contact = sort { $a cmp $b } find_records @key;
6218               my @contact = sort +find_records(@key);
6219               my @contact = sort &find_records(@key);
6220               my @contact = sort(find_records(@key));
6221
6222           If instead you want to sort the array @key with the comparison
6223           routine "find_records()" then you can use:
6224
6225               my @contact = sort { find_records() } @key;
6226               my @contact = sort find_records(@key);
6227               my @contact = sort(find_records @key);
6228               my @contact = sort(find_records (@key));
6229
6230           $a and $b are set as package globals in the package the sort() is
6231           called from.  That means $main::a and $main::b (or $::a and $::b)
6232           in the "main" package, $FooPack::a and $FooPack::b in the "FooPack"
6233           package, etc.  If the sort block is in scope of a "my" or "state"
6234           declaration of $a and/or $b, you must spell out the full name of
6235           the variables in the sort block :
6236
6237              package main;
6238              my $a = "C"; # DANGER, Will Robinson, DANGER !!!
6239
6240              print sort { $a cmp $b }               qw(A C E G B D F H);
6241                                                     # WRONG
6242              sub badlexi { $a cmp $b }
6243              print sort badlexi                     qw(A C E G B D F H);
6244                                                     # WRONG
6245              # the above prints BACFEDGH or some other incorrect ordering
6246
6247              print sort { $::a cmp $::b }           qw(A C E G B D F H);
6248                                                     # OK
6249              print sort { our $a cmp our $b }       qw(A C E G B D F H);
6250                                                     # also OK
6251              print sort { our ($a, $b); $a cmp $b } qw(A C E G B D F H);
6252                                                     # also OK
6253              sub lexi { our $a cmp our $b }
6254              print sort lexi                        qw(A C E G B D F H);
6255                                                     # also OK
6256              # the above print ABCDEFGH
6257
6258           With proper care you may mix package and my (or state) $a and/or
6259           $b:
6260
6261              my $a = {
6262                 tiny   => -2,
6263                 small  => -1,
6264                 normal => 0,
6265                 big    => 1,
6266                 huge   => 2
6267              };
6268
6269              say sort { $a->{our $a} <=> $a->{our $b} }
6270                  qw{ huge normal tiny small big};
6271
6272              # prints tinysmallnormalbighuge
6273
6274           $a and $b are implicitly local to the sort() execution and regain
6275           their former values upon completing the sort.
6276
6277           Sort subroutines written using $a and $b are bound to their calling
6278           package. It is possible, but of limited interest, to define them in
6279           a different package, since the subroutine must still refer to the
6280           calling package's $a and $b :
6281
6282              package Foo;
6283              sub lexi { $Bar::a cmp $Bar::b }
6284              package Bar;
6285              ... sort Foo::lexi ...
6286
6287           Use the prototyped versions (see above) for a more generic
6288           alternative.
6289
6290           The comparison function is required to behave.  If it returns
6291           inconsistent results (sometimes saying $x[1] is less than $x[2] and
6292           sometimes saying the opposite, for example) the results are not
6293           well-defined.
6294
6295           Because "<=>" returns "undef" when either operand is "NaN" (not-a-
6296           number), be careful when sorting with a comparison function like
6297           "$a <=> $b" any lists that might contain a "NaN".  The following
6298           example takes advantage that "NaN != NaN" to eliminate any "NaN"s
6299           from the input list.
6300
6301               my @result = sort { $a <=> $b } grep { $_ == $_ } @input;
6302
6303       splice ARRAY,OFFSET,LENGTH,LIST
6304       splice ARRAY,OFFSET,LENGTH
6305       splice ARRAY,OFFSET
6306       splice ARRAY
6307           Removes the elements designated by OFFSET and LENGTH from an array,
6308           and replaces them with the elements of LIST, if any.  In list
6309           context, returns the elements removed from the array.  In scalar
6310           context, returns the last element removed, or "undef" if no
6311           elements are removed.  The array grows or shrinks as necessary.  If
6312           OFFSET is negative then it starts that far from the end of the
6313           array.  If LENGTH is omitted, removes everything from OFFSET
6314           onward.  If LENGTH is negative, removes the elements from OFFSET
6315           onward except for -LENGTH elements at the end of the array.  If
6316           both OFFSET and LENGTH are omitted, removes everything.  If OFFSET
6317           is past the end of the array and a LENGTH was provided, Perl issues
6318           a warning, and splices at the end of the array.
6319
6320           The following equivalences hold (assuming "$#a >= $i" )
6321
6322               push(@a,$x,$y)      splice(@a,@a,0,$x,$y)
6323               pop(@a)             splice(@a,-1)
6324               shift(@a)           splice(@a,0,1)
6325               unshift(@a,$x,$y)   splice(@a,0,0,$x,$y)
6326               $a[$i] = $y         splice(@a,$i,1,$y)
6327
6328           "splice" can be used, for example, to implement n-ary queue
6329           processing:
6330
6331               sub nary_print {
6332                 my $n = shift;
6333                 while (my @next_n = splice @_, 0, $n) {
6334                   say join q{ -- }, @next_n;
6335                 }
6336               }
6337
6338               nary_print(3, qw(a b c d e f g h));
6339               # prints:
6340               #   a -- b -- c
6341               #   d -- e -- f
6342               #   g -- h
6343
6344           Starting with Perl 5.14, an experimental feature allowed "splice"
6345           to take a scalar expression. This experiment has been deemed
6346           unsuccessful, and was removed as of Perl 5.24.
6347
6348       split /PATTERN/,EXPR,LIMIT
6349       split /PATTERN/,EXPR
6350       split /PATTERN/
6351       split
6352           Splits the string EXPR into a list of strings and returns the list
6353           in list context, or the size of the list in scalar context.  (Prior
6354           to Perl 5.11, it also overwrote @_ with the list in void and scalar
6355           context. If you target old perls, beware.)
6356
6357           If only PATTERN is given, EXPR defaults to $_.
6358
6359           Anything in EXPR that matches PATTERN is taken to be a separator
6360           that separates the EXPR into substrings (called "fields") that do
6361           not include the separator.  Note that a separator may be longer
6362           than one character or even have no characters at all (the empty
6363           string, which is a zero-width match).
6364
6365           The PATTERN need not be constant; an expression may be used to
6366           specify a pattern that varies at runtime.
6367
6368           If PATTERN matches the empty string, the EXPR is split at the match
6369           position (between characters).  As an example, the following:
6370
6371               print join(':', split(/b/, 'abc')), "\n";
6372
6373           uses the "b" in 'abc' as a separator to produce the output "a:c".
6374           However, this:
6375
6376               print join(':', split(//, 'abc')), "\n";
6377
6378           uses empty string matches as separators to produce the output
6379           "a:b:c"; thus, the empty string may be used to split EXPR into a
6380           list of its component characters.
6381
6382           As a special case for "split", the empty pattern given in match
6383           operator syntax ("//") specifically matches the empty string, which
6384           is contrary to its usual interpretation as the last successful
6385           match.
6386
6387           If PATTERN is "/^/", then it is treated as if it used the multiline
6388           modifier ("/^/m"), since it isn't much use otherwise.
6389
6390           "/m" and any of the other pattern modifiers valid for "qr"
6391           (summarized in "qr/STRING/msixpodualn" in perlop) may be specified
6392           explicitly.
6393
6394           As another special case, "split" emulates the default behavior of
6395           the command line tool awk when the PATTERN is either omitted or a
6396           string composed of a single space character (such as ' ' or "\x20",
6397           but not e.g. "/ /").  In this case, any leading whitespace in EXPR
6398           is removed before splitting occurs, and the PATTERN is instead
6399           treated as if it were "/\s+/"; in particular, this means that any
6400           contiguous whitespace (not just a single space character) is used
6401           as a separator.  However, this special treatment can be avoided by
6402           specifying the pattern "/ /" instead of the string " ", thereby
6403           allowing only a single space character to be a separator.  In
6404           earlier Perls this special case was restricted to the use of a
6405           plain " " as the pattern argument to split; in Perl 5.18.0 and
6406           later this special case is triggered by any expression which
6407           evaluates to the simple string " ".
6408
6409           As of Perl 5.28, this special-cased whitespace splitting works as
6410           expected in the scope of "use feature 'unicode_strings". In
6411           previous versions, and outside the scope of that feature, it
6412           exhibits "The "Unicode Bug"" in perlunicode: characters that are
6413           whitespace according to Unicode rules but not according to ASCII
6414           rules can be treated as part of fields rather than as field
6415           separators, depending on the string's internal encoding.
6416
6417           If omitted, PATTERN defaults to a single space, " ", triggering the
6418           previously described awk emulation.
6419
6420           If LIMIT is specified and positive, it represents the maximum
6421           number of fields into which the EXPR may be split; in other words,
6422           LIMIT is one greater than the maximum number of times EXPR may be
6423           split.  Thus, the LIMIT value 1 means that EXPR may be split a
6424           maximum of zero times, producing a maximum of one field (namely,
6425           the entire value of EXPR).  For instance:
6426
6427               print join(':', split(//, 'abc', 1)), "\n";
6428
6429           produces the output "abc", and this:
6430
6431               print join(':', split(//, 'abc', 2)), "\n";
6432
6433           produces the output "a:bc", and each of these:
6434
6435               print join(':', split(//, 'abc', 3)), "\n";
6436               print join(':', split(//, 'abc', 4)), "\n";
6437
6438           produces the output "a:b:c".
6439
6440           If LIMIT is negative, it is treated as if it were instead
6441           arbitrarily large; as many fields as possible are produced.
6442
6443           If LIMIT is omitted (or, equivalently, zero), then it is usually
6444           treated as if it were instead negative but with the exception that
6445           trailing empty fields are stripped (empty leading fields are always
6446           preserved); if all fields are empty, then all fields are considered
6447           to be trailing (and are thus stripped in this case).  Thus, the
6448           following:
6449
6450               print join(':', split(/,/, 'a,b,c,,,')), "\n";
6451
6452           produces the output "a:b:c", but the following:
6453
6454               print join(':', split(/,/, 'a,b,c,,,', -1)), "\n";
6455
6456           produces the output "a:b:c:::".
6457
6458           In time-critical applications, it is worthwhile to avoid splitting
6459           into more fields than necessary.  Thus, when assigning to a list,
6460           if LIMIT is omitted (or zero), then LIMIT is treated as though it
6461           were one larger than the number of variables in the list; for the
6462           following, LIMIT is implicitly 3:
6463
6464               my ($login, $passwd) = split(/:/);
6465
6466           Note that splitting an EXPR that evaluates to the empty string
6467           always produces zero fields, regardless of the LIMIT specified.
6468
6469           An empty leading field is produced when there is a positive-width
6470           match at the beginning of EXPR.  For instance:
6471
6472               print join(':', split(/ /, ' abc')), "\n";
6473
6474           produces the output ":abc".  However, a zero-width match at the
6475           beginning of EXPR never produces an empty field, so that:
6476
6477               print join(':', split(//, ' abc'));
6478
6479           produces the output " :a:b:c" (rather than ": :a:b:c").
6480
6481           An empty trailing field, on the other hand, is produced when there
6482           is a match at the end of EXPR, regardless of the length of the
6483           match (of course, unless a non-zero LIMIT is given explicitly, such
6484           fields are removed, as in the last example).  Thus:
6485
6486               print join(':', split(//, ' abc', -1)), "\n";
6487
6488           produces the output " :a:b:c:".
6489
6490           If the PATTERN contains capturing groups, then for each separator,
6491           an additional field is produced for each substring captured by a
6492           group (in the order in which the groups are specified, as per
6493           backreferences); if any group does not match, then it captures the
6494           "undef" value instead of a substring.  Also, note that any such
6495           additional field is produced whenever there is a separator (that
6496           is, whenever a split occurs), and such an additional field does not
6497           count towards the LIMIT.  Consider the following expressions
6498           evaluated in list context (each returned list is provided in the
6499           associated comment):
6500
6501               split(/-|,/, "1-10,20", 3)
6502               # ('1', '10', '20')
6503
6504               split(/(-|,)/, "1-10,20", 3)
6505               # ('1', '-', '10', ',', '20')
6506
6507               split(/-|(,)/, "1-10,20", 3)
6508               # ('1', undef, '10', ',', '20')
6509
6510               split(/(-)|,/, "1-10,20", 3)
6511               # ('1', '-', '10', undef, '20')
6512
6513               split(/(-)|(,)/, "1-10,20", 3)
6514               # ('1', '-', undef, '10', undef, ',', '20')
6515
6516       sprintf FORMAT, LIST
6517           Returns a string formatted by the usual "printf" conventions of the
6518           C library function "sprintf".  See below for more details and see
6519           sprintf(3) or printf(3) on your system for an explanation of the
6520           general principles.
6521
6522           For example:
6523
6524                   # Format number with up to 8 leading zeroes
6525                   my $result = sprintf("%08d", $number);
6526
6527                   # Round number to 3 digits after decimal point
6528                   my $rounded = sprintf("%.3f", $number);
6529
6530           Perl does its own "sprintf" formatting: it emulates the C function
6531           sprintf(3), but doesn't use it except for floating-point numbers,
6532           and even then only standard modifiers are allowed.  Non-standard
6533           extensions in your local sprintf(3) are therefore unavailable from
6534           Perl.
6535
6536           Unlike "printf", "sprintf" does not do what you probably mean when
6537           you pass it an array as your first argument.  The array is given
6538           scalar context, and instead of using the 0th element of the array
6539           as the format, Perl will use the count of elements in the array as
6540           the format, which is almost never useful.
6541
6542           Perl's "sprintf" permits the following universally-known
6543           conversions:
6544
6545              %%    a percent sign
6546              %c    a character with the given number
6547              %s    a string
6548              %d    a signed integer, in decimal
6549              %u    an unsigned integer, in decimal
6550              %o    an unsigned integer, in octal
6551              %x    an unsigned integer, in hexadecimal
6552              %e    a floating-point number, in scientific notation
6553              %f    a floating-point number, in fixed decimal notation
6554              %g    a floating-point number, in %e or %f notation
6555
6556           In addition, Perl permits the following widely-supported
6557           conversions:
6558
6559              %X    like %x, but using upper-case letters
6560              %E    like %e, but using an upper-case "E"
6561              %G    like %g, but with an upper-case "E" (if applicable)
6562              %b    an unsigned integer, in binary
6563              %B    like %b, but using an upper-case "B" with the # flag
6564              %p    a pointer (outputs the Perl value's address in hexadecimal)
6565              %n    special: *stores* the number of characters output so far
6566                    into the next argument in the parameter list
6567              %a    hexadecimal floating point
6568              %A    like %a, but using upper-case letters
6569
6570           Finally, for backward (and we do mean "backward") compatibility,
6571           Perl permits these unnecessary but widely-supported conversions:
6572
6573              %i    a synonym for %d
6574              %D    a synonym for %ld
6575              %U    a synonym for %lu
6576              %O    a synonym for %lo
6577              %F    a synonym for %f
6578
6579           Note that the number of exponent digits in the scientific notation
6580           produced by %e, %E, %g and %G for numbers with the modulus of the
6581           exponent less than 100 is system-dependent: it may be three or less
6582           (zero-padded as necessary).  In other words, 1.23 times ten to the
6583           99th may be either "1.23e99" or "1.23e099".  Similarly for %a and
6584           %A: the exponent or the hexadecimal digits may float: especially
6585           the "long doubles" Perl configuration option may cause surprises.
6586
6587           Between the "%" and the format letter, you may specify several
6588           additional attributes controlling the interpretation of the format.
6589           In order, these are:
6590
6591           format parameter index
6592               An explicit format parameter index, such as "2$".  By default
6593               sprintf will format the next unused argument in the list, but
6594               this allows you to take the arguments out of order:
6595
6596                 printf '%2$d %1$d', 12, 34;      # prints "34 12"
6597                 printf '%3$d %d %1$d', 1, 2, 3;  # prints "3 1 1"
6598
6599           flags
6600               one or more of:
6601
6602                  space   prefix non-negative number with a space
6603                  +       prefix non-negative number with a plus sign
6604                  -       left-justify within the field
6605                  0       use zeros, not spaces, to right-justify
6606                  #       ensure the leading "0" for any octal,
6607                          prefix non-zero hexadecimal with "0x" or "0X",
6608                          prefix non-zero binary with "0b" or "0B"
6609
6610               For example:
6611
6612                 printf '<% d>',  12;   # prints "< 12>"
6613                 printf '<% d>',   0;   # prints "< 0>"
6614                 printf '<% d>', -12;   # prints "<-12>"
6615                 printf '<%+d>',  12;   # prints "<+12>"
6616                 printf '<%+d>',   0;   # prints "<+0>"
6617                 printf '<%+d>', -12;   # prints "<-12>"
6618                 printf '<%6s>',  12;   # prints "<    12>"
6619                 printf '<%-6s>', 12;   # prints "<12    >"
6620                 printf '<%06s>', 12;   # prints "<000012>"
6621                 printf '<%#o>',  12;   # prints "<014>"
6622                 printf '<%#x>',  12;   # prints "<0xc>"
6623                 printf '<%#X>',  12;   # prints "<0XC>"
6624                 printf '<%#b>',  12;   # prints "<0b1100>"
6625                 printf '<%#B>',  12;   # prints "<0B1100>"
6626
6627               When a space and a plus sign are given as the flags at once,
6628               the space is ignored.
6629
6630                 printf '<%+ d>', 12;   # prints "<+12>"
6631                 printf '<% +d>', 12;   # prints "<+12>"
6632
6633               When the # flag and a precision are given in the %o conversion,
6634               the precision is incremented if it's necessary for the leading
6635               "0".
6636
6637                 printf '<%#.5o>', 012;      # prints "<00012>"
6638                 printf '<%#.5o>', 012345;   # prints "<012345>"
6639                 printf '<%#.0o>', 0;        # prints "<0>"
6640
6641           vector flag
6642               This flag tells Perl to interpret the supplied string as a
6643               vector of integers, one for each character in the string.  Perl
6644               applies the format to each integer in turn, then joins the
6645               resulting strings with a separator (a dot "." by default).
6646               This can be useful for displaying ordinal values of characters
6647               in arbitrary strings:
6648
6649                 printf "%vd", "AB\x{100}";           # prints "65.66.256"
6650                 printf "version is v%vd\n", $^V;     # Perl's version
6651
6652               Put an asterisk "*" before the "v" to override the string to
6653               use to separate the numbers:
6654
6655                 printf "address is %*vX\n", ":", $addr;   # IPv6 address
6656                 printf "bits are %0*v8b\n", " ", $bits;   # random bitstring
6657
6658               You can also explicitly specify the argument number to use for
6659               the join string using something like "*2$v"; for example:
6660
6661                 printf '%*4$vX %*4$vX %*4$vX',       # 3 IPv6 addresses
6662                         @addr[1..3], ":";
6663
6664           (minimum) width
6665               Arguments are usually formatted to be only as wide as required
6666               to display the given value.  You can override the width by
6667               putting a number here, or get the width from the next argument
6668               (with "*") or from a specified argument (e.g., with "*2$"):
6669
6670                printf "<%s>", "a";       # prints "<a>"
6671                printf "<%6s>", "a";      # prints "<     a>"
6672                printf "<%*s>", 6, "a";   # prints "<     a>"
6673                printf '<%*2$s>', "a", 6; # prints "<     a>"
6674                printf "<%2s>", "long";   # prints "<long>" (does not truncate)
6675
6676               If a field width obtained through "*" is negative, it has the
6677               same effect as the "-" flag: left-justification.
6678
6679           precision, or maximum width
6680               You can specify a precision (for numeric conversions) or a
6681               maximum width (for string conversions) by specifying a "."
6682               followed by a number.  For floating-point formats except "g"
6683               and "G", this specifies how many places right of the decimal
6684               point to show (the default being 6).  For example:
6685
6686                 # these examples are subject to system-specific variation
6687                 printf '<%f>', 1;    # prints "<1.000000>"
6688                 printf '<%.1f>', 1;  # prints "<1.0>"
6689                 printf '<%.0f>', 1;  # prints "<1>"
6690                 printf '<%e>', 10;   # prints "<1.000000e+01>"
6691                 printf '<%.1e>', 10; # prints "<1.0e+01>"
6692
6693               For "g" and "G", this specifies the maximum number of
6694               significant digits to show; for example:
6695
6696                 # These examples are subject to system-specific variation.
6697                 printf '<%g>', 1;        # prints "<1>"
6698                 printf '<%.10g>', 1;     # prints "<1>"
6699                 printf '<%g>', 100;      # prints "<100>"
6700                 printf '<%.1g>', 100;    # prints "<1e+02>"
6701                 printf '<%.2g>', 100.01; # prints "<1e+02>"
6702                 printf '<%.5g>', 100.01; # prints "<100.01>"
6703                 printf '<%.4g>', 100.01; # prints "<100>"
6704                 printf '<%.1g>', 0.0111; # prints "<0.01>"
6705                 printf '<%.2g>', 0.0111; # prints "<0.011>"
6706                 printf '<%.3g>', 0.0111; # prints "<0.0111>"
6707
6708               For integer conversions, specifying a precision implies that
6709               the output of the number itself should be zero-padded to this
6710               width, where the 0 flag is ignored:
6711
6712                 printf '<%.6d>', 1;      # prints "<000001>"
6713                 printf '<%+.6d>', 1;     # prints "<+000001>"
6714                 printf '<%-10.6d>', 1;   # prints "<000001    >"
6715                 printf '<%10.6d>', 1;    # prints "<    000001>"
6716                 printf '<%010.6d>', 1;   # prints "<    000001>"
6717                 printf '<%+10.6d>', 1;   # prints "<   +000001>"
6718
6719                 printf '<%.6x>', 1;      # prints "<000001>"
6720                 printf '<%#.6x>', 1;     # prints "<0x000001>"
6721                 printf '<%-10.6x>', 1;   # prints "<000001    >"
6722                 printf '<%10.6x>', 1;    # prints "<    000001>"
6723                 printf '<%010.6x>', 1;   # prints "<    000001>"
6724                 printf '<%#10.6x>', 1;   # prints "<  0x000001>"
6725
6726               For string conversions, specifying a precision truncates the
6727               string to fit the specified width:
6728
6729                 printf '<%.5s>', "truncated";   # prints "<trunc>"
6730                 printf '<%10.5s>', "truncated"; # prints "<     trunc>"
6731
6732               You can also get the precision from the next argument using
6733               ".*", or from a specified argument (e.g., with ".*2$"):
6734
6735                 printf '<%.6x>', 1;       # prints "<000001>"
6736                 printf '<%.*x>', 6, 1;    # prints "<000001>"
6737
6738                 printf '<%.*2$x>', 1, 6;  # prints "<000001>"
6739
6740                 printf '<%6.*2$x>', 1, 4; # prints "<  0001>"
6741
6742               If a precision obtained through "*" is negative, it counts as
6743               having no precision at all.
6744
6745                 printf '<%.*s>',  7, "string";   # prints "<string>"
6746                 printf '<%.*s>',  3, "string";   # prints "<str>"
6747                 printf '<%.*s>',  0, "string";   # prints "<>"
6748                 printf '<%.*s>', -1, "string";   # prints "<string>"
6749
6750                 printf '<%.*d>',  1, 0;   # prints "<0>"
6751                 printf '<%.*d>',  0, 0;   # prints "<>"
6752                 printf '<%.*d>', -1, 0;   # prints "<0>"
6753
6754           size
6755               For numeric conversions, you can specify the size to interpret
6756               the number as using "l", "h", "V", "q", "L", or "ll".  For
6757               integer conversions ("d u o x X b i D U O"), numbers are
6758               usually assumed to be whatever the default integer size is on
6759               your platform (usually 32 or 64 bits), but you can override
6760               this to use instead one of the standard C types, as supported
6761               by the compiler used to build Perl:
6762
6763                  hh          interpret integer as C type "char" or "unsigned
6764                              char" on Perl 5.14 or later
6765                  h           interpret integer as C type "short" or
6766                              "unsigned short"
6767                  j           interpret integer as C type "intmax_t" on Perl
6768                              5.14 or later; and prior to Perl 5.30, only with
6769                              a C99 compiler (unportable)
6770                  l           interpret integer as C type "long" or
6771                              "unsigned long"
6772                  q, L, or ll interpret integer as C type "long long",
6773                              "unsigned long long", or "quad" (typically
6774                              64-bit integers)
6775                  t           interpret integer as C type "ptrdiff_t" on Perl
6776                              5.14 or later
6777                  z           interpret integer as C types "size_t" or
6778                              "ssize_t" on Perl 5.14 or later
6779
6780               As of 5.14, none of these raises an exception if they are not
6781               supported on your platform.  However, if warnings are enabled,
6782               a warning of the "printf" warning class is issued on an
6783               unsupported conversion flag.  Should you instead prefer an
6784               exception, do this:
6785
6786                   use warnings FATAL => "printf";
6787
6788               If you would like to know about a version dependency before you
6789               start running the program, put something like this at its top:
6790
6791                   use 5.014;  # for hh/j/t/z/ printf modifiers
6792
6793               You can find out whether your Perl supports quads via Config:
6794
6795                   use Config;
6796                   if ($Config{use64bitint} eq "define"
6797                       || $Config{longsize} >= 8) {
6798                       print "Nice quads!\n";
6799                   }
6800
6801               For floating-point conversions ("e f g E F G"), numbers are
6802               usually assumed to be the default floating-point size on your
6803               platform (double or long double), but you can force "long
6804               double" with "q", "L", or "ll" if your platform supports them.
6805               You can find out whether your Perl supports long doubles via
6806               Config:
6807
6808                   use Config;
6809                   print "long doubles\n" if $Config{d_longdbl} eq "define";
6810
6811               You can find out whether Perl considers "long double" to be the
6812               default floating-point size to use on your platform via Config:
6813
6814                   use Config;
6815                   if ($Config{uselongdouble} eq "define") {
6816                       print "long doubles by default\n";
6817                   }
6818
6819               It can also be that long doubles and doubles are the same
6820               thing:
6821
6822                       use Config;
6823                       ($Config{doublesize} == $Config{longdblsize}) &&
6824                               print "doubles are long doubles\n";
6825
6826               The size specifier "V" has no effect for Perl code, but is
6827               supported for compatibility with XS code.  It means "use the
6828               standard size for a Perl integer or floating-point number",
6829               which is the default.
6830
6831           order of arguments
6832               Normally, "sprintf" takes the next unused argument as the value
6833               to format for each format specification.  If the format
6834               specification uses "*" to require additional arguments, these
6835               are consumed from the argument list in the order they appear in
6836               the format specification before the value to format.  Where an
6837               argument is specified by an explicit index, this does not
6838               affect the normal order for the arguments, even when the
6839               explicitly specified index would have been the next argument.
6840
6841               So:
6842
6843                   printf "<%*.*s>", $a, $b, $c;
6844
6845               uses $a for the width, $b for the precision, and $c as the
6846               value to format; while:
6847
6848                 printf '<%*1$.*s>', $a, $b;
6849
6850               would use $a for the width and precision, and $b as the value
6851               to format.
6852
6853               Here are some more examples; be aware that when using an
6854               explicit index, the "$" may need escaping:
6855
6856                printf "%2\$d %d\n",      12, 34;     # will print "34 12\n"
6857                printf "%2\$d %d %d\n",   12, 34;     # will print "34 12 34\n"
6858                printf "%3\$d %d %d\n",   12, 34, 56; # will print "56 12 34\n"
6859                printf "%2\$*3\$d %d\n",  12, 34,  3; # will print " 34 12\n"
6860                printf "%*1\$.*f\n",       4,  5, 10; # will print "5.0000\n"
6861
6862           If "use locale" (including "use locale ':not_characters'") is in
6863           effect and "POSIX::setlocale" has been called, the character used
6864           for the decimal separator in formatted floating-point numbers is
6865           affected by the "LC_NUMERIC" locale.  See perllocale and POSIX.
6866
6867       sqrt EXPR
6868       sqrt
6869           Return the positive square root of EXPR.  If EXPR is omitted, uses
6870           $_.  Works only for non-negative operands unless you've loaded the
6871           "Math::Complex" module.
6872
6873               use Math::Complex;
6874               print sqrt(-4);    # prints 2i
6875
6876       srand EXPR
6877       srand
6878           Sets and returns the random number seed for the "rand" operator.
6879
6880           The point of the function is to "seed" the "rand" function so that
6881           "rand" can produce a different sequence each time you run your
6882           program.  When called with a parameter, "srand" uses that for the
6883           seed; otherwise it (semi-)randomly chooses a seed.  In either case,
6884           starting with Perl 5.14, it returns the seed.  To signal that your
6885           code will work only on Perls of a recent vintage:
6886
6887               use 5.014;  # so srand returns the seed
6888
6889           If "srand" is not called explicitly, it is called implicitly
6890           without a parameter at the first use of the "rand" operator.
6891           However, there are a few situations where programs are likely to
6892           want to call "srand".  One is for generating predictable results,
6893           generally for testing or debugging.  There, you use "srand($seed)",
6894           with the same $seed each time.  Another case is that you may want
6895           to call "srand" after a "fork" to avoid child processes sharing the
6896           same seed value as the parent (and consequently each other).
6897
6898           Do not call "srand()" (i.e., without an argument) more than once
6899           per process.  The internal state of the random number generator
6900           should contain more entropy than can be provided by any seed, so
6901           calling "srand" again actually loses randomness.
6902
6903           Most implementations of "srand" take an integer and will silently
6904           truncate decimal numbers.  This means "srand(42)" will usually
6905           produce the same results as "srand(42.1)".  To be safe, always pass
6906           "srand" an integer.
6907
6908           A typical use of the returned seed is for a test program which has
6909           too many combinations to test comprehensively in the time available
6910           to it each run.  It can test a random subset each time, and should
6911           there be a failure, log the seed used for that run so that it can
6912           later be used to reproduce the same results.
6913
6914           "rand" is not cryptographically secure.  You should not rely on it
6915           in security-sensitive situations.  As of this writing, a number of
6916           third-party CPAN modules offer random number generators intended by
6917           their authors to be cryptographically secure, including:
6918           Data::Entropy, Crypt::Random, Math::Random::Secure, and
6919           Math::TrulyRandom.
6920
6921       stat FILEHANDLE
6922       stat EXPR
6923       stat DIRHANDLE
6924       stat
6925           Returns a 13-element list giving the status info for a file, either
6926           the file opened via FILEHANDLE or DIRHANDLE, or named by EXPR.  If
6927           EXPR is omitted, it stats $_ (not "_"!).  Returns the empty list if
6928           "stat" fails.  Typically used as follows:
6929
6930               my ($dev,$ino,$mode,$nlink,$uid,$gid,$rdev,$size,
6931                   $atime,$mtime,$ctime,$blksize,$blocks)
6932                      = stat($filename);
6933
6934           Not all fields are supported on all filesystem types.  Here are the
6935           meanings of the fields:
6936
6937             0 dev      device number of filesystem
6938             1 ino      inode number
6939             2 mode     file mode  (type and permissions)
6940             3 nlink    number of (hard) links to the file
6941             4 uid      numeric user ID of file's owner
6942             5 gid      numeric group ID of file's owner
6943             6 rdev     the device identifier (special files only)
6944             7 size     total size of file, in bytes
6945             8 atime    last access time in seconds since the epoch
6946             9 mtime    last modify time in seconds since the epoch
6947            10 ctime    inode change time in seconds since the epoch (*)
6948            11 blksize  preferred I/O size in bytes for interacting with the
6949                        file (may vary from file to file)
6950            12 blocks   actual number of system-specific blocks allocated
6951                        on disk (often, but not always, 512 bytes each)
6952
6953           (The epoch was at 00:00 January 1, 1970 GMT.)
6954
6955           (*) Not all fields are supported on all filesystem types.  Notably,
6956           the ctime field is non-portable.  In particular, you cannot expect
6957           it to be a "creation time"; see "Files and Filesystems" in perlport
6958           for details.
6959
6960           If "stat" is passed the special filehandle consisting of an
6961           underline, no stat is done, but the current contents of the stat
6962           structure from the last "stat", "lstat", or filetest are returned.
6963           Example:
6964
6965               if (-x $file && (($d) = stat(_)) && $d < 0) {
6966                   print "$file is executable NFS file\n";
6967               }
6968
6969           (This works on machines only for which the device number is
6970           negative under NFS.)
6971
6972           On some platforms inode numbers are of a type larger than perl
6973           knows how to handle as integer numerical values.  If necessary, an
6974           inode number will be returned as a decimal string in order to
6975           preserve the entire value.  If used in a numeric context, this will
6976           be converted to a floating-point numerical value, with rounding, a
6977           fate that is best avoided.  Therefore, you should prefer to compare
6978           inode numbers using "eq" rather than "==".  "eq" will work fine on
6979           inode numbers that are represented numerically, as well as those
6980           represented as strings.
6981
6982           Because the mode contains both the file type and its permissions,
6983           you should mask off the file type portion and (s)printf using a
6984           "%o" if you want to see the real permissions.
6985
6986               my $mode = (stat($filename))[2];
6987               printf "Permissions are %04o\n", $mode & 07777;
6988
6989           In scalar context, "stat" returns a boolean value indicating
6990           success or failure, and, if successful, sets the information
6991           associated with the special filehandle "_".
6992
6993           The File::stat module provides a convenient, by-name access
6994           mechanism:
6995
6996               use File::stat;
6997               my $sb = stat($filename);
6998               printf "File is %s, size is %s, perm %04o, mtime %s\n",
6999                      $filename, $sb->size, $sb->mode & 07777,
7000                      scalar localtime $sb->mtime;
7001
7002           You can import symbolic mode constants ("S_IF*") and functions
7003           ("S_IS*") from the Fcntl module:
7004
7005               use Fcntl ':mode';
7006
7007               my $mode = (stat($filename))[2];
7008
7009               my $user_rwx      = ($mode & S_IRWXU) >> 6;
7010               my $group_read    = ($mode & S_IRGRP) >> 3;
7011               my $other_execute =  $mode & S_IXOTH;
7012
7013               printf "Permissions are %04o\n", S_IMODE($mode), "\n";
7014
7015               my $is_setuid     =  $mode & S_ISUID;
7016               my $is_directory  =  S_ISDIR($mode);
7017
7018           You could write the last two using the "-u" and "-d" operators.
7019           Commonly available "S_IF*" constants are:
7020
7021               # Permissions: read, write, execute, for user, group, others.
7022
7023               S_IRWXU S_IRUSR S_IWUSR S_IXUSR
7024               S_IRWXG S_IRGRP S_IWGRP S_IXGRP
7025               S_IRWXO S_IROTH S_IWOTH S_IXOTH
7026
7027               # Setuid/Setgid/Stickiness/SaveText.
7028               # Note that the exact meaning of these is system-dependent.
7029
7030               S_ISUID S_ISGID S_ISVTX S_ISTXT
7031
7032               # File types.  Not all are necessarily available on
7033               # your system.
7034
7035               S_IFREG S_IFDIR S_IFLNK S_IFBLK S_IFCHR
7036               S_IFIFO S_IFSOCK S_IFWHT S_ENFMT
7037
7038               # The following are compatibility aliases for S_IRUSR,
7039               # S_IWUSR, and S_IXUSR.
7040
7041               S_IREAD S_IWRITE S_IEXEC
7042
7043           and the "S_IF*" functions are
7044
7045               S_IMODE($mode)    the part of $mode containing the permission
7046                                 bits and the setuid/setgid/sticky bits
7047
7048               S_IFMT($mode)     the part of $mode containing the file type
7049                                 which can be bit-anded with (for example)
7050                                 S_IFREG or with the following functions
7051
7052               # The operators -f, -d, -l, -b, -c, -p, and -S.
7053
7054               S_ISREG($mode) S_ISDIR($mode) S_ISLNK($mode)
7055               S_ISBLK($mode) S_ISCHR($mode) S_ISFIFO($mode) S_ISSOCK($mode)
7056
7057               # No direct -X operator counterpart, but for the first one
7058               # the -g operator is often equivalent.  The ENFMT stands for
7059               # record flocking enforcement, a platform-dependent feature.
7060
7061               S_ISENFMT($mode) S_ISWHT($mode)
7062
7063           See your native chmod(2) and stat(2) documentation for more details
7064           about the "S_*" constants.  To get status info for a symbolic link
7065           instead of the target file behind the link, use the "lstat"
7066           function.
7067
7068           Portability issues: "stat" in perlport.
7069
7070       state VARLIST
7071       state TYPE VARLIST
7072       state VARLIST : ATTRS
7073       state TYPE VARLIST : ATTRS
7074           "state" declares a lexically scoped variable, just like "my".
7075           However, those variables will never be reinitialized, contrary to
7076           lexical variables that are reinitialized each time their enclosing
7077           block is entered.  See "Persistent Private Variables" in perlsub
7078           for details.
7079
7080           If more than one variable is listed, the list must be placed in
7081           parentheses.  With a parenthesised list, "undef" can be used as a
7082           dummy placeholder.  However, since initialization of state
7083           variables in such lists is currently not possible this would serve
7084           no purpose.
7085
7086           "state" is available only if the "state" feature is enabled or if
7087           it is prefixed with "CORE::".  The "state" feature is enabled
7088           automatically with a "use v5.10" (or higher) declaration in the
7089           current scope.
7090
7091       study SCALAR
7092       study
7093           At this time, "study" does nothing. This may change in the future.
7094
7095           Prior to Perl version 5.16, it would create an inverted index of
7096           all characters that occurred in the given SCALAR (or $_ if
7097           unspecified). When matching a pattern, the rarest character from
7098           the pattern would be looked up in this index. Rarity was based on
7099           some static frequency tables constructed from some C programs and
7100           English text.
7101
7102       sub NAME BLOCK
7103       sub NAME (PROTO) BLOCK
7104       sub NAME : ATTRS BLOCK
7105       sub NAME (PROTO) : ATTRS BLOCK
7106           This is subroutine definition, not a real function per se.  Without
7107           a BLOCK it's just a forward declaration.  Without a NAME, it's an
7108           anonymous function declaration, so does return a value: the CODE
7109           ref of the closure just created.
7110
7111           See perlsub and perlref for details about subroutines and
7112           references; see attributes and Attribute::Handlers for more
7113           information about attributes.
7114
7115       __SUB__
7116           A special token that returns a reference to the current subroutine,
7117           or "undef" outside of a subroutine.
7118
7119           The behaviour of "__SUB__" within a regex code block (such as
7120           "/(?{...})/") is subject to change.
7121
7122           This token is only available under "use v5.16" or the "current_sub"
7123           feature.  See feature.
7124
7125       substr EXPR,OFFSET,LENGTH,REPLACEMENT
7126       substr EXPR,OFFSET,LENGTH
7127       substr EXPR,OFFSET
7128           Extracts a substring out of EXPR and returns it.  First character
7129           is at offset zero.  If OFFSET is negative, starts that far back
7130           from the end of the string.  If LENGTH is omitted, returns
7131           everything through the end of the string.  If LENGTH is negative,
7132           leaves that many characters off the end of the string.
7133
7134               my $s = "The black cat climbed the green tree";
7135               my $color  = substr $s, 4, 5;      # black
7136               my $middle = substr $s, 4, -11;    # black cat climbed the
7137               my $end    = substr $s, 14;        # climbed the green tree
7138               my $tail   = substr $s, -4;        # tree
7139               my $z      = substr $s, -4, 2;     # tr
7140
7141           You can use the "substr" function as an lvalue, in which case EXPR
7142           must itself be an lvalue.  If you assign something shorter than
7143           LENGTH, the string will shrink, and if you assign something longer
7144           than LENGTH, the string will grow to accommodate it.  To keep the
7145           string the same length, you may need to pad or chop your value
7146           using "sprintf".
7147
7148           If OFFSET and LENGTH specify a substring that is partly outside the
7149           string, only the part within the string is returned.  If the
7150           substring is beyond either end of the string, "substr" returns the
7151           undefined value and produces a warning.  When used as an lvalue,
7152           specifying a substring that is entirely outside the string raises
7153           an exception.  Here's an example showing the behavior for boundary
7154           cases:
7155
7156               my $name = 'fred';
7157               substr($name, 4) = 'dy';         # $name is now 'freddy'
7158               my $null = substr $name, 6, 2;   # returns "" (no warning)
7159               my $oops = substr $name, 7;      # returns undef, with warning
7160               substr($name, 7) = 'gap';        # raises an exception
7161
7162           An alternative to using "substr" as an lvalue is to specify the
7163           replacement string as the 4th argument.  This allows you to replace
7164           parts of the EXPR and return what was there before in one
7165           operation, just as you can with "splice".
7166
7167               my $s = "The black cat climbed the green tree";
7168               my $z = substr $s, 14, 7, "jumped from";    # climbed
7169               # $s is now "The black cat jumped from the green tree"
7170
7171           Note that the lvalue returned by the three-argument version of
7172           "substr" acts as a 'magic bullet'; each time it is assigned to, it
7173           remembers which part of the original string is being modified; for
7174           example:
7175
7176               my $x = '1234';
7177               for (substr($x,1,2)) {
7178                   $_ = 'a';   print $x,"\n";    # prints 1a4
7179                   $_ = 'xyz'; print $x,"\n";    # prints 1xyz4
7180                   $x = '56789';
7181                   $_ = 'pq';  print $x,"\n";    # prints 5pq9
7182               }
7183
7184           With negative offsets, it remembers its position from the end of
7185           the string when the target string is modified:
7186
7187               my $x = '1234';
7188               for (substr($x, -3, 2)) {
7189                   $_ = 'a';   print $x,"\n";    # prints 1a4, as above
7190                   $x = 'abcdefg';
7191                   print $_,"\n";                # prints f
7192               }
7193
7194           Prior to Perl version 5.10, the result of using an lvalue multiple
7195           times was unspecified.  Prior to 5.16, the result with negative
7196           offsets was unspecified.
7197
7198       symlink OLDFILE,NEWFILE
7199           Creates a new filename symbolically linked to the old filename.
7200           Returns 1 for success, 0 otherwise.  On systems that don't support
7201           symbolic links, raises an exception.  To check for that, use eval:
7202
7203               my $symlink_exists = eval { symlink("",""); 1 };
7204
7205           Portability issues: "symlink" in perlport.
7206
7207       syscall NUMBER, LIST
7208           Calls the system call specified as the first element of the list,
7209           passing the remaining elements as arguments to the system call.  If
7210           unimplemented, raises an exception.  The arguments are interpreted
7211           as follows: if a given argument is numeric, the argument is passed
7212           as an int.  If not, the pointer to the string value is passed.  You
7213           are responsible to make sure a string is pre-extended long enough
7214           to receive any result that might be written into a string.  You
7215           can't use a string literal (or other read-only string) as an
7216           argument to "syscall" because Perl has to assume that any string
7217           pointer might be written through.  If your integer arguments are
7218           not literals and have never been interpreted in a numeric context,
7219           you may need to add 0 to them to force them to look like numbers.
7220           This emulates the "syswrite" function (or vice versa):
7221
7222               require 'syscall.ph';        # may need to run h2ph
7223               my $s = "hi there\n";
7224               syscall(SYS_write(), fileno(STDOUT), $s, length $s);
7225
7226           Note that Perl supports passing of up to only 14 arguments to your
7227           syscall, which in practice should (usually) suffice.
7228
7229           Syscall returns whatever value returned by the system call it
7230           calls.  If the system call fails, "syscall" returns "-1" and sets
7231           $! (errno).  Note that some system calls can legitimately return
7232           "-1".  The proper way to handle such calls is to assign "$! = 0"
7233           before the call, then check the value of $! if "syscall" returns
7234           "-1".
7235
7236           There's a problem with "syscall(SYS_pipe())": it returns the file
7237           number of the read end of the pipe it creates, but there is no way
7238           to retrieve the file number of the other end.  You can avoid this
7239           problem by using "pipe" instead.
7240
7241           Portability issues: "syscall" in perlport.
7242
7243       sysopen FILEHANDLE,FILENAME,MODE
7244       sysopen FILEHANDLE,FILENAME,MODE,PERMS
7245           Opens the file whose filename is given by FILENAME, and associates
7246           it with FILEHANDLE.  If FILEHANDLE is an expression, its value is
7247           used as the real filehandle wanted; an undefined scalar will be
7248           suitably autovivified.  This function calls the underlying
7249           operating system's open(2) function with the parameters FILENAME,
7250           MODE, and PERMS.
7251
7252           Returns true on success and "undef" otherwise.
7253
7254           PerlIO layers will be applied to the handle the same way they would
7255           in an "open" call that does not specify layers. That is, the
7256           current value of "${^OPEN}" as set by the open pragma in a lexical
7257           scope, or the "-C" commandline option or "PERL_UNICODE" environment
7258           variable in the main program scope, falling back to the platform
7259           defaults as described in "Defaults and how to override them" in
7260           PerlIO. If you want to remove any layers that may transform the
7261           byte stream, use "binmode" after opening it.
7262
7263           The possible values and flag bits of the MODE parameter are system-
7264           dependent; they are available via the standard module "Fcntl".  See
7265           the documentation of your operating system's open(2) syscall to see
7266           which values and flag bits are available.  You may combine several
7267           flags using the "|"-operator.
7268
7269           Some of the most common values are "O_RDONLY" for opening the file
7270           in read-only mode, "O_WRONLY" for opening the file in write-only
7271           mode, and "O_RDWR" for opening the file in read-write mode.
7272
7273           For historical reasons, some values work on almost every system
7274           supported by Perl: 0 means read-only, 1 means write-only, and 2
7275           means read/write.  We know that these values do not work under
7276           OS/390 and on the Macintosh; you probably don't want to use them in
7277           new code.
7278
7279           If the file named by FILENAME does not exist and the "open" call
7280           creates it (typically because MODE includes the "O_CREAT" flag),
7281           then the value of PERMS specifies the permissions of the newly
7282           created file.  If you omit the PERMS argument to "sysopen", Perl
7283           uses the octal value 0666.  These permission values need to be in
7284           octal, and are modified by your process's current "umask".
7285
7286           In many systems the "O_EXCL" flag is available for opening files in
7287           exclusive mode.  This is not locking: exclusiveness means here that
7288           if the file already exists, "sysopen" fails.  "O_EXCL" may not work
7289           on network filesystems, and has no effect unless the "O_CREAT" flag
7290           is set as well.  Setting "O_CREAT|O_EXCL" prevents the file from
7291           being opened if it is a symbolic link.  It does not protect against
7292           symbolic links in the file's path.
7293
7294           Sometimes you may want to truncate an already-existing file.  This
7295           can be done using the "O_TRUNC" flag.  The behavior of "O_TRUNC"
7296           with "O_RDONLY" is undefined.
7297
7298           You should seldom if ever use 0644 as argument to "sysopen",
7299           because that takes away the user's option to have a more permissive
7300           umask.  Better to omit it.  See "umask" for more on this.
7301
7302           This function has no direct relation to the usage of "sysread",
7303           "syswrite", or "sysseek".  A handle opened with this function can
7304           be used with buffered IO just as one opened with "open" can be used
7305           with unbuffered IO.
7306
7307           Note that under Perls older than 5.8.0, "sysopen" depends on the
7308           fdopen(3) C library function.  On many Unix systems, fdopen(3) is
7309           known to fail when file descriptors exceed a certain value,
7310           typically 255.  If you need more file descriptors than that,
7311           consider using the "POSIX::open" function.  For Perls 5.8.0 and
7312           later, PerlIO is (most often) the default.
7313
7314           See perlopentut for a kinder, gentler explanation of opening files.
7315
7316           Portability issues: "sysopen" in perlport.
7317
7318       sysread FILEHANDLE,SCALAR,LENGTH,OFFSET
7319       sysread FILEHANDLE,SCALAR,LENGTH
7320           Attempts to read LENGTH bytes of data into variable SCALAR from the
7321           specified FILEHANDLE, using read(2).  It bypasses any PerlIO layers
7322           including buffered IO (but is affected by the presence of the
7323           ":utf8" layer as described later), so mixing this with other kinds
7324           of reads, "print", "write", "seek", "tell", or "eof" can cause
7325           confusion because the ":perlio" or ":crlf" layers usually buffer
7326           data.  Returns the number of bytes actually read, 0 at end of file,
7327           or undef if there was an error (in the latter case $! is also set).
7328           SCALAR will be grown or shrunk so that the last byte actually read
7329           is the last byte of the scalar after the read.
7330
7331           An OFFSET may be specified to place the read data at some place in
7332           the string other than the beginning.  A negative OFFSET specifies
7333           placement at that many characters counting backwards from the end
7334           of the string.  A positive OFFSET greater than the length of SCALAR
7335           results in the string being padded to the required size with "\0"
7336           bytes before the result of the read is appended.
7337
7338           There is no syseof() function, which is ok, since "eof" doesn't
7339           work well on device files (like ttys) anyway.  Use "sysread" and
7340           check for a return value of 0 to decide whether you're done.
7341
7342           Note that if the filehandle has been marked as ":utf8", "sysread"
7343           will throw an exception.  The ":encoding(...)" layer implicitly
7344           introduces the ":utf8" layer.  See "binmode", "open", and the open
7345           pragma.
7346
7347       sysseek FILEHANDLE,POSITION,WHENCE
7348           Sets FILEHANDLE's system position in bytes using lseek(2).
7349           FILEHANDLE may be an expression whose value gives the name of the
7350           filehandle.  The values for WHENCE are 0 to set the new position to
7351           POSITION; 1 to set it to the current position plus POSITION; and 2
7352           to set it to EOF plus POSITION, typically negative.
7353
7354           Note the emphasis on bytes: even if the filehandle has been set to
7355           operate on characters (for example using the ":encoding(UTF-8)" I/O
7356           layer), the "seek", "tell", and "sysseek" family of functions use
7357           byte offsets, not character offsets, because seeking to a character
7358           offset would be very slow in a UTF-8 file.
7359
7360           "sysseek" bypasses normal buffered IO, so mixing it with reads
7361           other than "sysread" (for example "readline" or "read"), "print",
7362           "write", "seek", "tell", or "eof" may cause confusion.
7363
7364           For WHENCE, you may also use the constants "SEEK_SET", "SEEK_CUR",
7365           and "SEEK_END" (start of the file, current position, end of the
7366           file) from the Fcntl module.  Use of the constants is also more
7367           portable than relying on 0, 1, and 2.  For example to define a
7368           "systell" function:
7369
7370               use Fcntl 'SEEK_CUR';
7371               sub systell { sysseek($_[0], 0, SEEK_CUR) }
7372
7373           Returns the new position, or the undefined value on failure.  A
7374           position of zero is returned as the string "0 but true"; thus
7375           "sysseek" returns true on success and false on failure, yet you can
7376           still easily determine the new position.
7377
7378       system LIST
7379       system PROGRAM LIST
7380           Does exactly the same thing as "exec", except that a fork is done
7381           first and the parent process waits for the child process to exit.
7382           Note that argument processing varies depending on the number of
7383           arguments.  If there is more than one argument in LIST, or if LIST
7384           is an array with more than one value, starts the program given by
7385           the first element of the list with arguments given by the rest of
7386           the list.  If there is only one scalar argument, the argument is
7387           checked for shell metacharacters, and if there are any, the entire
7388           argument is passed to the system's command shell for parsing (this
7389           is "/bin/sh -c" on Unix platforms, but varies on other platforms).
7390           If there are no shell metacharacters in the argument, it is split
7391           into words and passed directly to "execvp", which is more
7392           efficient.  On Windows, only the "system PROGRAM LIST" syntax will
7393           reliably avoid using the shell; "system LIST", even with more than
7394           one element, will fall back to the shell if the first spawn fails.
7395
7396           Perl will attempt to flush all files opened for output before any
7397           operation that may do a fork, but this may not be supported on some
7398           platforms (see perlport).  To be safe, you may need to set $|
7399           ($AUTOFLUSH in English) or call the "autoflush" method of
7400           "IO::Handle" on any open handles.
7401
7402           The return value is the exit status of the program as returned by
7403           the "wait" call.  To get the actual exit value, shift right by
7404           eight (see below).  See also "exec".  This is not what you want to
7405           use to capture the output from a command; for that you should use
7406           merely backticks or "qx//", as described in "`STRING`" in perlop.
7407           Return value of -1 indicates a failure to start the program or an
7408           error of the wait(2) system call (inspect $! for the reason).
7409
7410           If you'd like to make "system" (and many other bits of Perl) die on
7411           error, have a look at the autodie pragma.
7412
7413           Like "exec", "system" allows you to lie to a program about its name
7414           if you use the "system PROGRAM LIST" syntax.  Again, see "exec".
7415
7416           Since "SIGINT" and "SIGQUIT" are ignored during the execution of
7417           "system", if you expect your program to terminate on receipt of
7418           these signals you will need to arrange to do so yourself based on
7419           the return value.
7420
7421               my @args = ("command", "arg1", "arg2");
7422               system(@args) == 0
7423                   or die "system @args failed: $?";
7424
7425           If you'd like to manually inspect "system"'s failure, you can check
7426           all possible failure modes by inspecting $? like this:
7427
7428               if ($? == -1) {
7429                   print "failed to execute: $!\n";
7430               }
7431               elsif ($? & 127) {
7432                   printf "child died with signal %d, %s coredump\n",
7433                       ($? & 127),  ($? & 128) ? 'with' : 'without';
7434               }
7435               else {
7436                   printf "child exited with value %d\n", $? >> 8;
7437               }
7438
7439           Alternatively, you may inspect the value of
7440           "${^CHILD_ERROR_NATIVE}" with the "W*()" calls from the POSIX
7441           module.
7442
7443           When "system"'s arguments are executed indirectly by the shell,
7444           results and return codes are subject to its quirks.  See "`STRING`"
7445           in perlop and "exec" for details.
7446
7447           Since "system" does a "fork" and "wait" it may affect a "SIGCHLD"
7448           handler.  See perlipc for details.
7449
7450           Portability issues: "system" in perlport.
7451
7452       syswrite FILEHANDLE,SCALAR,LENGTH,OFFSET
7453       syswrite FILEHANDLE,SCALAR,LENGTH
7454       syswrite FILEHANDLE,SCALAR
7455           Attempts to write LENGTH bytes of data from variable SCALAR to the
7456           specified FILEHANDLE, using write(2).  If LENGTH is not specified,
7457           writes whole SCALAR.  It bypasses any PerlIO layers including
7458           buffered IO (but is affected by the presence of the ":utf8" layer
7459           as described later), so mixing this with reads (other than
7460           "sysread)"), "print", "write", "seek", "tell", or "eof" may cause
7461           confusion because the ":perlio" and ":crlf" layers usually buffer
7462           data.  Returns the number of bytes actually written, or "undef" if
7463           there was an error (in this case the errno variable $! is also
7464           set).  If the LENGTH is greater than the data available in the
7465           SCALAR after the OFFSET, only as much data as is available will be
7466           written.
7467
7468           An OFFSET may be specified to write the data from some part of the
7469           string other than the beginning.  A negative OFFSET specifies
7470           writing that many characters counting backwards from the end of the
7471           string.  If SCALAR is of length zero, you can only use an OFFSET of
7472           0.
7473
7474           WARNING: If the filehandle is marked ":utf8", "syswrite" will raise
7475           an exception.  The ":encoding(...)" layer implicitly introduces the
7476           ":utf8" layer.  Alternately, if the handle is not marked with an
7477           encoding but you attempt to write characters with code points over
7478           255, raises an exception.  See "binmode", "open", and the open
7479           pragma.
7480
7481       tell FILEHANDLE
7482       tell
7483           Returns the current position in bytes for FILEHANDLE, or -1 on
7484           error.  FILEHANDLE may be an expression whose value gives the name
7485           of the actual filehandle.  If FILEHANDLE is omitted, assumes the
7486           file last read.
7487
7488           Note the emphasis on bytes: even if the filehandle has been set to
7489           operate on characters (for example using the ":encoding(UTF-8)" I/O
7490           layer), the "seek", "tell", and "sysseek" family of functions use
7491           byte offsets, not character offsets, because seeking to a character
7492           offset would be very slow in a UTF-8 file.
7493
7494           The return value of "tell" for the standard streams like the STDIN
7495           depends on the operating system: it may return -1 or something
7496           else.  "tell" on pipes, fifos, and sockets usually returns -1.
7497
7498           There is no "systell" function.  Use "sysseek($fh, 0, 1)" for that.
7499
7500           Do not use "tell" (or other buffered I/O operations) on a
7501           filehandle that has been manipulated by "sysread", "syswrite", or
7502           "sysseek".  Those functions ignore the buffering, while "tell" does
7503           not.
7504
7505       telldir DIRHANDLE
7506           Returns the current position of the "readdir" routines on
7507           DIRHANDLE.  Value may be given to "seekdir" to access a particular
7508           location in a directory.  "telldir" has the same caveats about
7509           possible directory compaction as the corresponding system library
7510           routine.
7511
7512       tie VARIABLE,CLASSNAME,LIST
7513           This function binds a variable to a package class that will provide
7514           the implementation for the variable.  VARIABLE is the name of the
7515           variable to be enchanted.  CLASSNAME is the name of a class
7516           implementing objects of correct type.  Any additional arguments are
7517           passed to the appropriate constructor method of the class (meaning
7518           "TIESCALAR", "TIEHANDLE", "TIEARRAY", or "TIEHASH").  Typically
7519           these are arguments such as might be passed to the dbm_open(3)
7520           function of C.  The object returned by the constructor is also
7521           returned by the "tie" function, which would be useful if you want
7522           to access other methods in CLASSNAME.
7523
7524           Note that functions such as "keys" and "values" may return huge
7525           lists when used on large objects, like DBM files.  You may prefer
7526           to use the "each" function to iterate over such.  Example:
7527
7528               # print out history file offsets
7529               use NDBM_File;
7530               tie(my %HIST, 'NDBM_File', '/usr/lib/news/history', 1, 0);
7531               while (my ($key,$val) = each %HIST) {
7532                   print $key, ' = ', unpack('L', $val), "\n";
7533               }
7534
7535           A class implementing a hash should have the following methods:
7536
7537               TIEHASH classname, LIST
7538               FETCH this, key
7539               STORE this, key, value
7540               DELETE this, key
7541               CLEAR this
7542               EXISTS this, key
7543               FIRSTKEY this
7544               NEXTKEY this, lastkey
7545               SCALAR this
7546               DESTROY this
7547               UNTIE this
7548
7549           A class implementing an ordinary array should have the following
7550           methods:
7551
7552               TIEARRAY classname, LIST
7553               FETCH this, key
7554               STORE this, key, value
7555               FETCHSIZE this
7556               STORESIZE this, count
7557               CLEAR this
7558               PUSH this, LIST
7559               POP this
7560               SHIFT this
7561               UNSHIFT this, LIST
7562               SPLICE this, offset, length, LIST
7563               EXTEND this, count
7564               DELETE this, key
7565               EXISTS this, key
7566               DESTROY this
7567               UNTIE this
7568
7569           A class implementing a filehandle should have the following
7570           methods:
7571
7572               TIEHANDLE classname, LIST
7573               READ this, scalar, length, offset
7574               READLINE this
7575               GETC this
7576               WRITE this, scalar, length, offset
7577               PRINT this, LIST
7578               PRINTF this, format, LIST
7579               BINMODE this
7580               EOF this
7581               FILENO this
7582               SEEK this, position, whence
7583               TELL this
7584               OPEN this, mode, LIST
7585               CLOSE this
7586               DESTROY this
7587               UNTIE this
7588
7589           A class implementing a scalar should have the following methods:
7590
7591               TIESCALAR classname, LIST
7592               FETCH this,
7593               STORE this, value
7594               DESTROY this
7595               UNTIE this
7596
7597           Not all methods indicated above need be implemented.  See perltie,
7598           Tie::Hash, Tie::Array, Tie::Scalar, and Tie::Handle.
7599
7600           Unlike "dbmopen", the "tie" function will not "use" or "require" a
7601           module for you; you need to do that explicitly yourself.  See
7602           DB_File or the Config module for interesting "tie" implementations.
7603
7604           For further details see perltie, "tied".
7605
7606       tied VARIABLE
7607           Returns a reference to the object underlying VARIABLE (the same
7608           value that was originally returned by the "tie" call that bound the
7609           variable to a package.)  Returns the undefined value if VARIABLE
7610           isn't tied to a package.
7611
7612       time
7613           Returns the number of non-leap seconds since whatever time the
7614           system considers to be the epoch, suitable for feeding to "gmtime"
7615           and "localtime".  On most systems the epoch is 00:00:00 UTC,
7616           January 1, 1970; a prominent exception being Mac OS Classic which
7617           uses 00:00:00, January 1, 1904 in the current local time zone for
7618           its epoch.
7619
7620           For measuring time in better granularity than one second, use the
7621           Time::HiRes module from Perl 5.8 onwards (or from CPAN before
7622           then), or, if you have gettimeofday(2), you may be able to use the
7623           "syscall" interface of Perl.  See perlfaq8 for details.
7624
7625           For date and time processing look at the many related modules on
7626           CPAN.  For a comprehensive date and time representation look at the
7627           DateTime module.
7628
7629       times
7630           Returns a four-element list giving the user and system times in
7631           seconds for this process and any exited children of this process.
7632
7633               my ($user,$system,$cuser,$csystem) = times;
7634
7635           In scalar context, "times" returns $user.
7636
7637           Children's times are only included for terminated children.
7638
7639           Portability issues: "times" in perlport.
7640
7641       tr///
7642           The transliteration operator.  Same as "y///".  See "Quote-Like
7643           Operators" in perlop.
7644
7645       truncate FILEHANDLE,LENGTH
7646       truncate EXPR,LENGTH
7647           Truncates the file opened on FILEHANDLE, or named by EXPR, to the
7648           specified length.  Raises an exception if truncate isn't
7649           implemented on your system.  Returns true if successful, "undef" on
7650           error.
7651
7652           The behavior is undefined if LENGTH is greater than the length of
7653           the file.
7654
7655           The position in the file of FILEHANDLE is left unchanged.  You may
7656           want to call seek before writing to the file.
7657
7658           Portability issues: "truncate" in perlport.
7659
7660       uc EXPR
7661       uc  Returns an uppercased version of EXPR.  This is the internal
7662           function implementing the "\U" escape in double-quoted strings.  It
7663           does not attempt to do titlecase mapping on initial letters.  See
7664           "ucfirst" for that.
7665
7666           If EXPR is omitted, uses $_.
7667
7668           This function behaves the same way under various pragmas, such as
7669           in a locale, as "lc" does.
7670
7671       ucfirst EXPR
7672       ucfirst
7673           Returns the value of EXPR with the first character in uppercase
7674           (titlecase in Unicode).  This is the internal function implementing
7675           the "\u" escape in double-quoted strings.
7676
7677           If EXPR is omitted, uses $_.
7678
7679           This function behaves the same way under various pragmas, such as
7680           in a locale, as "lc" does.
7681
7682       umask EXPR
7683       umask
7684           Sets the umask for the process to EXPR and returns the previous
7685           value.  If EXPR is omitted, merely returns the current umask.
7686
7687           The Unix permission "rwxr-x---" is represented as three sets of
7688           three bits, or three octal digits: 0750 (the leading 0 indicates
7689           octal and isn't one of the digits).  The "umask" value is such a
7690           number representing disabled permissions bits.  The permission (or
7691           "mode") values you pass "mkdir" or "sysopen" are modified by your
7692           umask, so even if you tell "sysopen" to create a file with
7693           permissions 0777, if your umask is 0022, then the file will
7694           actually be created with permissions 0755.  If your "umask" were
7695           0027 (group can't write; others can't read, write, or execute),
7696           then passing "sysopen" 0666 would create a file with mode 0640
7697           (because "0666 &~ 027" is 0640).
7698
7699           Here's some advice: supply a creation mode of 0666 for regular
7700           files (in "sysopen") and one of 0777 for directories (in "mkdir")
7701           and executable files.  This gives users the freedom of choice: if
7702           they want protected files, they might choose process umasks of 022,
7703           027, or even the particularly antisocial mask of 077.  Programs
7704           should rarely if ever make policy decisions better left to the
7705           user.  The exception to this is when writing files that should be
7706           kept private: mail files, web browser cookies, .rhosts files, and
7707           so on.
7708
7709           If umask(2) is not implemented on your system and you are trying to
7710           restrict access for yourself (i.e., "(EXPR & 0700) > 0"), raises an
7711           exception.  If umask(2) is not implemented and you are not trying
7712           to restrict access for yourself, returns "undef".
7713
7714           Remember that a umask is a number, usually given in octal; it is
7715           not a string of octal digits.  See also "oct", if all you have is a
7716           string.
7717
7718           Portability issues: "umask" in perlport.
7719
7720       undef EXPR
7721       undef
7722           Undefines the value of EXPR, which must be an lvalue.  Use only on
7723           a scalar value, an array (using "@"), a hash (using "%"), a
7724           subroutine (using "&"), or a typeglob (using "*").  Saying "undef
7725           $hash{$key}" will probably not do what you expect on most
7726           predefined variables or DBM list values, so don't do that; see
7727           "delete".  Always returns the undefined value.  You can omit the
7728           EXPR, in which case nothing is undefined, but you still get an
7729           undefined value that you could, for instance, return from a
7730           subroutine, assign to a variable, or pass as a parameter.
7731           Examples:
7732
7733               undef $foo;
7734               undef $bar{'blurfl'};      # Compare to: delete $bar{'blurfl'};
7735               undef @ary;
7736               undef %hash;
7737               undef &mysub;
7738               undef *xyz;       # destroys $xyz, @xyz, %xyz, &xyz, etc.
7739               return (wantarray ? (undef, $errmsg) : undef) if $they_blew_it;
7740               select undef, undef, undef, 0.25;
7741               my ($x, $y, undef, $z) = foo();    # Ignore third value returned
7742
7743           Note that this is a unary operator, not a list operator.
7744
7745       unlink LIST
7746       unlink
7747           Deletes a list of files.  On success, it returns the number of
7748           files it successfully deleted.  On failure, it returns false and
7749           sets $! (errno):
7750
7751               my $unlinked = unlink 'a', 'b', 'c';
7752               unlink @goners;
7753               unlink glob "*.bak";
7754
7755           On error, "unlink" will not tell you which files it could not
7756           remove.  If you want to know which files you could not remove, try
7757           them one at a time:
7758
7759                foreach my $file ( @goners ) {
7760                    unlink $file or warn "Could not unlink $file: $!";
7761                }
7762
7763           Note: "unlink" will not attempt to delete directories unless you
7764           are superuser and the -U flag is supplied to Perl.  Even if these
7765           conditions are met, be warned that unlinking a directory can
7766           inflict damage on your filesystem.  Finally, using "unlink" on
7767           directories is not supported on many operating systems.  Use
7768           "rmdir" instead.
7769
7770           If LIST is omitted, "unlink" uses $_.
7771
7772       unpack TEMPLATE,EXPR
7773       unpack TEMPLATE
7774           "unpack" does the reverse of "pack": it takes a string and expands
7775           it out into a list of values.  (In scalar context, it returns
7776           merely the first value produced.)
7777
7778           If EXPR is omitted, unpacks the $_ string.  See perlpacktut for an
7779           introduction to this function.
7780
7781           The string is broken into chunks described by the TEMPLATE.  Each
7782           chunk is converted separately to a value.  Typically, either the
7783           string is a result of "pack", or the characters of the string
7784           represent a C structure of some kind.
7785
7786           The TEMPLATE has the same format as in the "pack" function.  Here's
7787           a subroutine that does substring:
7788
7789               sub substr {
7790                   my ($what, $where, $howmuch) = @_;
7791                   unpack("x$where a$howmuch", $what);
7792               }
7793
7794           and then there's
7795
7796               sub ordinal { unpack("W",$_[0]); } # same as ord()
7797
7798           In addition to fields allowed in "pack", you may prefix a field
7799           with a %<number> to indicate that you want a <number>-bit checksum
7800           of the items instead of the items themselves.  Default is a 16-bit
7801           checksum.  The checksum is calculated by summing numeric values of
7802           expanded values (for string fields the sum of "ord($char)" is
7803           taken; for bit fields the sum of zeroes and ones).
7804
7805           For example, the following computes the same number as the System V
7806           sum program:
7807
7808               my $checksum = do {
7809                   local $/;  # slurp!
7810                   unpack("%32W*", readline) % 65535;
7811               };
7812
7813           The following efficiently counts the number of set bits in a bit
7814           vector:
7815
7816               my $setbits = unpack("%32b*", $selectmask);
7817
7818           The "p" and "P" formats should be used with care.  Since Perl has
7819           no way of checking whether the value passed to "unpack" corresponds
7820           to a valid memory location, passing a pointer value that's not
7821           known to be valid is likely to have disastrous consequences.
7822
7823           If there are more pack codes or if the repeat count of a field or a
7824           group is larger than what the remainder of the input string allows,
7825           the result is not well defined: the repeat count may be decreased,
7826           or "unpack" may produce empty strings or zeros, or it may raise an
7827           exception.  If the input string is longer than one described by the
7828           TEMPLATE, the remainder of that input string is ignored.
7829
7830           See "pack" for more examples and notes.
7831
7832       unshift ARRAY,LIST
7833           Does the opposite of a "shift".  Or the opposite of a "push",
7834           depending on how you look at it.  Prepends list to the front of the
7835           array and returns the new number of elements in the array.
7836
7837               unshift(@ARGV, '-e') unless $ARGV[0] =~ /^-/;
7838
7839           Note the LIST is prepended whole, not one element at a time, so the
7840           prepended elements stay in the same order.  Use "reverse" to do the
7841           reverse.
7842
7843           Starting with Perl 5.14, an experimental feature allowed "unshift"
7844           to take a scalar expression. This experiment has been deemed
7845           unsuccessful, and was removed as of Perl 5.24.
7846
7847       untie VARIABLE
7848           Breaks the binding between a variable and a package.  (See tie.)
7849           Has no effect if the variable is not tied.
7850
7851       use Module VERSION LIST
7852       use Module VERSION
7853       use Module LIST
7854       use Module
7855       use VERSION
7856           Imports some semantics into the current package from the named
7857           module, generally by aliasing certain subroutine or variable names
7858           into your package.  It is exactly equivalent to
7859
7860               BEGIN { require Module; Module->import( LIST ); }
7861
7862           except that Module must be a bareword.  The importation can be made
7863           conditional by using the if module.
7864
7865           In the "use VERSION" form, VERSION may be either a v-string such as
7866           v5.24.1, which will be compared to $^V (aka $PERL_VERSION), or a
7867           numeric argument of the form 5.024001, which will be compared to
7868           $].  An exception is raised if VERSION is greater than the version
7869           of the current Perl interpreter; Perl will not attempt to parse the
7870           rest of the file.  Compare with "require", which can do a similar
7871           check at run time.  Symmetrically, "no VERSION" allows you to
7872           specify that you want a version of Perl older than the specified
7873           one.
7874
7875           Specifying VERSION as a numeric argument of the form 5.024001
7876           should generally be avoided as older less readable syntax compared
7877           to v5.24.1. Before perl 5.8.0 released in 2002 the more verbose
7878           numeric form was the only supported syntax, which is why you might
7879           see it in
7880
7881               use v5.24.1;    # compile time version check
7882               use 5.24.1;     # ditto
7883               use 5.024_001;  # ditto; older syntax compatible with perl 5.6
7884
7885           This is often useful if you need to check the current Perl version
7886           before "use"ing library modules that won't work with older versions
7887           of Perl.  (We try not to do this more than we have to.)
7888
7889           "use VERSION" also lexically enables all features available in the
7890           requested version as defined by the feature pragma, disabling any
7891           features not in the requested version's feature bundle.  See
7892           feature.  Similarly, if the specified Perl version is greater than
7893           or equal to 5.12.0, strictures are enabled lexically as with "use
7894           strict".  Any explicit use of "use strict" or "no strict" overrides
7895           "use VERSION", even if it comes before it.  Later use of "use
7896           VERSION" will override all behavior of a previous "use VERSION",
7897           possibly removing the "strict" and "feature" added by "use
7898           VERSION".  "use VERSION" does not load the feature.pm or strict.pm
7899           files.
7900
7901           The "BEGIN" forces the "require" and "import" to happen at compile
7902           time.  The "require" makes sure the module is loaded into memory if
7903           it hasn't been yet.  The "import" is not a builtin; it's just an
7904           ordinary static method call into the "Module" package to tell the
7905           module to import the list of features back into the current
7906           package.  The module can implement its "import" method any way it
7907           likes, though most modules just choose to derive their "import"
7908           method via inheritance from the "Exporter" class that is defined in
7909           the "Exporter" module.  See Exporter.  If no "import" method can be
7910           found, then the call is skipped, even if there is an AUTOLOAD
7911           method.
7912
7913           If you do not want to call the package's "import" method (for
7914           instance, to stop your namespace from being altered), explicitly
7915           supply the empty list:
7916
7917               use Module ();
7918
7919           That is exactly equivalent to
7920
7921               BEGIN { require Module }
7922
7923           If the VERSION argument is present between Module and LIST, then
7924           the "use" will call the "VERSION" method in class Module with the
7925           given version as an argument:
7926
7927               use Module 12.34;
7928
7929           is equivalent to:
7930
7931               BEGIN { require Module; Module->VERSION(12.34) }
7932
7933           The default "VERSION" method, inherited from the "UNIVERSAL" class,
7934           croaks if the given version is larger than the value of the
7935           variable $Module::VERSION.
7936
7937           The VERSION argument cannot be an arbitrary expression.  It only
7938           counts as a VERSION argument if it is a version number literal,
7939           starting with either a digit or "v" followed by a digit.  Anything
7940           that doesn't look like a version literal will be parsed as the
7941           start of the LIST.  Nevertheless, many attempts to use an arbitrary
7942           expression as a VERSION argument will appear to work, because
7943           Exporter's "import" method handles numeric arguments specially,
7944           performing version checks rather than treating them as things to
7945           export.
7946
7947           Again, there is a distinction between omitting LIST ("import"
7948           called with no arguments) and an explicit empty LIST "()" ("import"
7949           not called).  Note that there is no comma after VERSION!
7950
7951           Because this is a wide-open interface, pragmas (compiler
7952           directives) are also implemented this way.  Some of the currently
7953           implemented pragmas are:
7954
7955               use constant;
7956               use diagnostics;
7957               use integer;
7958               use sigtrap  qw(SEGV BUS);
7959               use strict   qw(subs vars refs);
7960               use subs     qw(afunc blurfl);
7961               use warnings qw(all);
7962               use sort     qw(stable);
7963
7964           Some of these pseudo-modules import semantics into the current
7965           block scope (like "strict" or "integer", unlike ordinary modules,
7966           which import symbols into the current package (which are effective
7967           through the end of the file).
7968
7969           Because "use" takes effect at compile time, it doesn't respect the
7970           ordinary flow control of the code being compiled.  In particular,
7971           putting a "use" inside the false branch of a conditional doesn't
7972           prevent it from being processed.  If a module or pragma only needs
7973           to be loaded conditionally, this can be done using the if pragma:
7974
7975               use if $] < 5.008, "utf8";
7976               use if WANT_WARNINGS, warnings => qw(all);
7977
7978           There's a corresponding "no" declaration that unimports meanings
7979           imported by "use", i.e., it calls "Module->unimport(LIST)" instead
7980           of "import".  It behaves just as "import" does with VERSION, an
7981           omitted or empty LIST, or no unimport method being found.
7982
7983               no integer;
7984               no strict 'refs';
7985               no warnings;
7986
7987           Care should be taken when using the "no VERSION" form of "no".  It
7988           is only meant to be used to assert that the running Perl is of a
7989           earlier version than its argument and not to undo the feature-
7990           enabling side effects of "use VERSION".
7991
7992           See perlmodlib for a list of standard modules and pragmas.  See
7993           perlrun for the "-M" and "-m" command-line options to Perl that
7994           give "use" functionality from the command-line.
7995
7996       utime LIST
7997           Changes the access and modification times on each file of a list of
7998           files.  The first two elements of the list must be the NUMERIC
7999           access and modification times, in that order.  Returns the number
8000           of files successfully changed.  The inode change time of each file
8001           is set to the current time.  For example, this code has the same
8002           effect as the Unix touch(1) command when the files already exist
8003           and belong to the user running the program:
8004
8005               #!/usr/bin/perl
8006               my $atime = my $mtime = time;
8007               utime $atime, $mtime, @ARGV;
8008
8009           Since Perl 5.8.0, if the first two elements of the list are
8010           "undef", the utime(2) syscall from your C library is called with a
8011           null second argument.  On most systems, this will set the file's
8012           access and modification times to the current time (i.e., equivalent
8013           to the example above) and will work even on files you don't own
8014           provided you have write permission:
8015
8016               for my $file (@ARGV) {
8017                   utime(undef, undef, $file)
8018                       || warn "Couldn't touch $file: $!";
8019               }
8020
8021           Under NFS this will use the time of the NFS server, not the time of
8022           the local machine.  If there is a time synchronization problem, the
8023           NFS server and local machine will have different times.  The Unix
8024           touch(1) command will in fact normally use this form instead of the
8025           one shown in the first example.
8026
8027           Passing only one of the first two elements as "undef" is equivalent
8028           to passing a 0 and will not have the effect described when both are
8029           "undef".  This also triggers an uninitialized warning.
8030
8031           On systems that support futimes(2), you may pass filehandles among
8032           the files.  On systems that don't support futimes(2), passing
8033           filehandles raises an exception.  Filehandles must be passed as
8034           globs or glob references to be recognized; barewords are considered
8035           filenames.
8036
8037           Portability issues: "utime" in perlport.
8038
8039       values HASH
8040       values ARRAY
8041           In list context, returns a list consisting of all the values of the
8042           named hash.  In Perl 5.12 or later only, will also return a list of
8043           the values of an array; prior to that release, attempting to use an
8044           array argument will produce a syntax error.  In scalar context,
8045           returns the number of values.
8046
8047           Hash entries are returned in an apparently random order.  The
8048           actual random order is specific to a given hash; the exact same
8049           series of operations on two hashes may result in a different order
8050           for each hash.  Any insertion into the hash may change the order,
8051           as will any deletion, with the exception that the most recent key
8052           returned by "each" or "keys" may be deleted without changing the
8053           order.  So long as a given hash is unmodified you may rely on
8054           "keys", "values" and "each" to repeatedly return the same order as
8055           each other.  See "Algorithmic Complexity Attacks" in perlsec for
8056           details on why hash order is randomized.  Aside from the guarantees
8057           provided here the exact details of Perl's hash algorithm and the
8058           hash traversal order are subject to change in any release of Perl.
8059           Tied hashes may behave differently to Perl's hashes with respect to
8060           changes in order on insertion and deletion of items.
8061
8062           As a side effect, calling "values" resets the HASH or ARRAY's
8063           internal iterator (see "each") before yielding the values.  In
8064           particular, calling "values" in void context resets the iterator
8065           with no other overhead.
8066
8067           Apart from resetting the iterator, "values @array" in list context
8068           is the same as plain @array.  (We recommend that you use void
8069           context "keys @array" for this, but reasoned that taking "values
8070           @array" out would require more documentation than leaving it in.)
8071
8072           Note that the values are not copied, which means modifying them
8073           will modify the contents of the hash:
8074
8075               for (values %hash)      { s/foo/bar/g }  # modifies %hash values
8076               for (@hash{keys %hash}) { s/foo/bar/g }  # same
8077
8078           Starting with Perl 5.14, an experimental feature allowed "values"
8079           to take a scalar expression. This experiment has been deemed
8080           unsuccessful, and was removed as of Perl 5.24.
8081
8082           To avoid confusing would-be users of your code who are running
8083           earlier versions of Perl with mysterious syntax errors, put this
8084           sort of thing at the top of your file to signal that your code will
8085           work only on Perls of a recent vintage:
8086
8087               use 5.012;  # so keys/values/each work on arrays
8088
8089           See also "keys", "each", and "sort".
8090
8091       vec EXPR,OFFSET,BITS
8092           Treats the string in EXPR as a bit vector made up of elements of
8093           width BITS and returns the value of the element specified by OFFSET
8094           as an unsigned integer.  BITS therefore specifies the number of
8095           bits that are reserved for each element in the bit vector.  This
8096           must be a power of two from 1 to 32 (or 64, if your platform
8097           supports that).
8098
8099           If BITS is 8, "elements" coincide with bytes of the input string.
8100
8101           If BITS is 16 or more, bytes of the input string are grouped into
8102           chunks of size BITS/8, and each group is converted to a number as
8103           with "pack"/"unpack" with big-endian formats "n"/"N" (and
8104           analogously for BITS==64).  See "pack" for details.
8105
8106           If bits is 4 or less, the string is broken into bytes, then the
8107           bits of each byte are broken into 8/BITS groups.  Bits of a byte
8108           are numbered in a little-endian-ish way, as in 0x01, 0x02, 0x04,
8109           0x08, 0x10, 0x20, 0x40, 0x80.  For example, breaking the single
8110           input byte "chr(0x36)" into two groups gives a list "(0x6, 0x3)";
8111           breaking it into 4 groups gives "(0x2, 0x1, 0x3, 0x0)".
8112
8113           "vec" may also be assigned to, in which case parentheses are needed
8114           to give the expression the correct precedence as in
8115
8116               vec($image, $max_x * $x + $y, 8) = 3;
8117
8118           If the selected element is outside the string, the value 0 is
8119           returned.  If an element off the end of the string is written to,
8120           Perl will first extend the string with sufficiently many zero
8121           bytes.   It is an error to try to write off the beginning of the
8122           string (i.e., negative OFFSET).
8123
8124           If the string happens to be encoded as UTF-8 internally (and thus
8125           has the UTF8 flag set), "vec" tries to convert it to use a one-
8126           byte-per-character internal representation. However, if the string
8127           contains characters with values of 256 or higher, a fatal error
8128           will occur.
8129
8130           Strings created with "vec" can also be manipulated with the logical
8131           operators "|", "&", "^", and "~".  These operators will assume a
8132           bit vector operation is desired when both operands are strings.
8133           See "Bitwise String Operators" in perlop.
8134
8135           The following code will build up an ASCII string saying
8136           'PerlPerlPerl'.  The comments show the string after each step.
8137           Note that this code works in the same way on big-endian or little-
8138           endian machines.
8139
8140               my $foo = '';
8141               vec($foo,  0, 32) = 0x5065726C; # 'Perl'
8142
8143               # $foo eq "Perl" eq "\x50\x65\x72\x6C", 32 bits
8144               print vec($foo, 0, 8);  # prints 80 == 0x50 == ord('P')
8145
8146               vec($foo,  2, 16) = 0x5065; # 'PerlPe'
8147               vec($foo,  3, 16) = 0x726C; # 'PerlPerl'
8148               vec($foo,  8,  8) = 0x50;   # 'PerlPerlP'
8149               vec($foo,  9,  8) = 0x65;   # 'PerlPerlPe'
8150               vec($foo, 20,  4) = 2;      # 'PerlPerlPe'   . "\x02"
8151               vec($foo, 21,  4) = 7;      # 'PerlPerlPer'
8152                                              # 'r' is "\x72"
8153               vec($foo, 45,  2) = 3;      # 'PerlPerlPer'  . "\x0c"
8154               vec($foo, 93,  1) = 1;      # 'PerlPerlPer'  . "\x2c"
8155               vec($foo, 94,  1) = 1;      # 'PerlPerlPerl'
8156                                              # 'l' is "\x6c"
8157
8158           To transform a bit vector into a string or list of 0's and 1's, use
8159           these:
8160
8161               my $bits = unpack("b*", $vector);
8162               my @bits = split(//, unpack("b*", $vector));
8163
8164           If you know the exact length in bits, it can be used in place of
8165           the "*".
8166
8167           Here is an example to illustrate how the bits actually fall in
8168           place:
8169
8170             #!/usr/bin/perl -wl
8171
8172             print <<'EOT';
8173                                               0         1         2         3
8174                                unpack("V",$_) 01234567890123456789012345678901
8175             ------------------------------------------------------------------
8176             EOT
8177
8178             for $w (0..3) {
8179                 $width = 2**$w;
8180                 for ($shift=0; $shift < $width; ++$shift) {
8181                     for ($off=0; $off < 32/$width; ++$off) {
8182                         $str = pack("B*", "0"x32);
8183                         $bits = (1<<$shift);
8184                         vec($str, $off, $width) = $bits;
8185                         $res = unpack("b*",$str);
8186                         $val = unpack("V", $str);
8187                         write;
8188                     }
8189                 }
8190             }
8191
8192             format STDOUT =
8193             vec($_,@#,@#) = @<< == @######### @>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
8194             $off, $width, $bits, $val, $res
8195             .
8196             __END__
8197
8198           Regardless of the machine architecture on which it runs, the
8199           example above should print the following table:
8200
8201                                               0         1         2         3
8202                                unpack("V",$_) 01234567890123456789012345678901
8203             ------------------------------------------------------------------
8204             vec($_, 0, 1) = 1   ==          1 10000000000000000000000000000000
8205             vec($_, 1, 1) = 1   ==          2 01000000000000000000000000000000
8206             vec($_, 2, 1) = 1   ==          4 00100000000000000000000000000000
8207             vec($_, 3, 1) = 1   ==          8 00010000000000000000000000000000
8208             vec($_, 4, 1) = 1   ==         16 00001000000000000000000000000000
8209             vec($_, 5, 1) = 1   ==         32 00000100000000000000000000000000
8210             vec($_, 6, 1) = 1   ==         64 00000010000000000000000000000000
8211             vec($_, 7, 1) = 1   ==        128 00000001000000000000000000000000
8212             vec($_, 8, 1) = 1   ==        256 00000000100000000000000000000000
8213             vec($_, 9, 1) = 1   ==        512 00000000010000000000000000000000
8214             vec($_,10, 1) = 1   ==       1024 00000000001000000000000000000000
8215             vec($_,11, 1) = 1   ==       2048 00000000000100000000000000000000
8216             vec($_,12, 1) = 1   ==       4096 00000000000010000000000000000000
8217             vec($_,13, 1) = 1   ==       8192 00000000000001000000000000000000
8218             vec($_,14, 1) = 1   ==      16384 00000000000000100000000000000000
8219             vec($_,15, 1) = 1   ==      32768 00000000000000010000000000000000
8220             vec($_,16, 1) = 1   ==      65536 00000000000000001000000000000000
8221             vec($_,17, 1) = 1   ==     131072 00000000000000000100000000000000
8222             vec($_,18, 1) = 1   ==     262144 00000000000000000010000000000000
8223             vec($_,19, 1) = 1   ==     524288 00000000000000000001000000000000
8224             vec($_,20, 1) = 1   ==    1048576 00000000000000000000100000000000
8225             vec($_,21, 1) = 1   ==    2097152 00000000000000000000010000000000
8226             vec($_,22, 1) = 1   ==    4194304 00000000000000000000001000000000
8227             vec($_,23, 1) = 1   ==    8388608 00000000000000000000000100000000
8228             vec($_,24, 1) = 1   ==   16777216 00000000000000000000000010000000
8229             vec($_,25, 1) = 1   ==   33554432 00000000000000000000000001000000
8230             vec($_,26, 1) = 1   ==   67108864 00000000000000000000000000100000
8231             vec($_,27, 1) = 1   ==  134217728 00000000000000000000000000010000
8232             vec($_,28, 1) = 1   ==  268435456 00000000000000000000000000001000
8233             vec($_,29, 1) = 1   ==  536870912 00000000000000000000000000000100
8234             vec($_,30, 1) = 1   == 1073741824 00000000000000000000000000000010
8235             vec($_,31, 1) = 1   == 2147483648 00000000000000000000000000000001
8236             vec($_, 0, 2) = 1   ==          1 10000000000000000000000000000000
8237             vec($_, 1, 2) = 1   ==          4 00100000000000000000000000000000
8238             vec($_, 2, 2) = 1   ==         16 00001000000000000000000000000000
8239             vec($_, 3, 2) = 1   ==         64 00000010000000000000000000000000
8240             vec($_, 4, 2) = 1   ==        256 00000000100000000000000000000000
8241             vec($_, 5, 2) = 1   ==       1024 00000000001000000000000000000000
8242             vec($_, 6, 2) = 1   ==       4096 00000000000010000000000000000000
8243             vec($_, 7, 2) = 1   ==      16384 00000000000000100000000000000000
8244             vec($_, 8, 2) = 1   ==      65536 00000000000000001000000000000000
8245             vec($_, 9, 2) = 1   ==     262144 00000000000000000010000000000000
8246             vec($_,10, 2) = 1   ==    1048576 00000000000000000000100000000000
8247             vec($_,11, 2) = 1   ==    4194304 00000000000000000000001000000000
8248             vec($_,12, 2) = 1   ==   16777216 00000000000000000000000010000000
8249             vec($_,13, 2) = 1   ==   67108864 00000000000000000000000000100000
8250             vec($_,14, 2) = 1   ==  268435456 00000000000000000000000000001000
8251             vec($_,15, 2) = 1   == 1073741824 00000000000000000000000000000010
8252             vec($_, 0, 2) = 2   ==          2 01000000000000000000000000000000
8253             vec($_, 1, 2) = 2   ==          8 00010000000000000000000000000000
8254             vec($_, 2, 2) = 2   ==         32 00000100000000000000000000000000
8255             vec($_, 3, 2) = 2   ==        128 00000001000000000000000000000000
8256             vec($_, 4, 2) = 2   ==        512 00000000010000000000000000000000
8257             vec($_, 5, 2) = 2   ==       2048 00000000000100000000000000000000
8258             vec($_, 6, 2) = 2   ==       8192 00000000000001000000000000000000
8259             vec($_, 7, 2) = 2   ==      32768 00000000000000010000000000000000
8260             vec($_, 8, 2) = 2   ==     131072 00000000000000000100000000000000
8261             vec($_, 9, 2) = 2   ==     524288 00000000000000000001000000000000
8262             vec($_,10, 2) = 2   ==    2097152 00000000000000000000010000000000
8263             vec($_,11, 2) = 2   ==    8388608 00000000000000000000000100000000
8264             vec($_,12, 2) = 2   ==   33554432 00000000000000000000000001000000
8265             vec($_,13, 2) = 2   ==  134217728 00000000000000000000000000010000
8266             vec($_,14, 2) = 2   ==  536870912 00000000000000000000000000000100
8267             vec($_,15, 2) = 2   == 2147483648 00000000000000000000000000000001
8268             vec($_, 0, 4) = 1   ==          1 10000000000000000000000000000000
8269             vec($_, 1, 4) = 1   ==         16 00001000000000000000000000000000
8270             vec($_, 2, 4) = 1   ==        256 00000000100000000000000000000000
8271             vec($_, 3, 4) = 1   ==       4096 00000000000010000000000000000000
8272             vec($_, 4, 4) = 1   ==      65536 00000000000000001000000000000000
8273             vec($_, 5, 4) = 1   ==    1048576 00000000000000000000100000000000
8274             vec($_, 6, 4) = 1   ==   16777216 00000000000000000000000010000000
8275             vec($_, 7, 4) = 1   ==  268435456 00000000000000000000000000001000
8276             vec($_, 0, 4) = 2   ==          2 01000000000000000000000000000000
8277             vec($_, 1, 4) = 2   ==         32 00000100000000000000000000000000
8278             vec($_, 2, 4) = 2   ==        512 00000000010000000000000000000000
8279             vec($_, 3, 4) = 2   ==       8192 00000000000001000000000000000000
8280             vec($_, 4, 4) = 2   ==     131072 00000000000000000100000000000000
8281             vec($_, 5, 4) = 2   ==    2097152 00000000000000000000010000000000
8282             vec($_, 6, 4) = 2   ==   33554432 00000000000000000000000001000000
8283             vec($_, 7, 4) = 2   ==  536870912 00000000000000000000000000000100
8284             vec($_, 0, 4) = 4   ==          4 00100000000000000000000000000000
8285             vec($_, 1, 4) = 4   ==         64 00000010000000000000000000000000
8286             vec($_, 2, 4) = 4   ==       1024 00000000001000000000000000000000
8287             vec($_, 3, 4) = 4   ==      16384 00000000000000100000000000000000
8288             vec($_, 4, 4) = 4   ==     262144 00000000000000000010000000000000
8289             vec($_, 5, 4) = 4   ==    4194304 00000000000000000000001000000000
8290             vec($_, 6, 4) = 4   ==   67108864 00000000000000000000000000100000
8291             vec($_, 7, 4) = 4   == 1073741824 00000000000000000000000000000010
8292             vec($_, 0, 4) = 8   ==          8 00010000000000000000000000000000
8293             vec($_, 1, 4) = 8   ==        128 00000001000000000000000000000000
8294             vec($_, 2, 4) = 8   ==       2048 00000000000100000000000000000000
8295             vec($_, 3, 4) = 8   ==      32768 00000000000000010000000000000000
8296             vec($_, 4, 4) = 8   ==     524288 00000000000000000001000000000000
8297             vec($_, 5, 4) = 8   ==    8388608 00000000000000000000000100000000
8298             vec($_, 6, 4) = 8   ==  134217728 00000000000000000000000000010000
8299             vec($_, 7, 4) = 8   == 2147483648 00000000000000000000000000000001
8300             vec($_, 0, 8) = 1   ==          1 10000000000000000000000000000000
8301             vec($_, 1, 8) = 1   ==        256 00000000100000000000000000000000
8302             vec($_, 2, 8) = 1   ==      65536 00000000000000001000000000000000
8303             vec($_, 3, 8) = 1   ==   16777216 00000000000000000000000010000000
8304             vec($_, 0, 8) = 2   ==          2 01000000000000000000000000000000
8305             vec($_, 1, 8) = 2   ==        512 00000000010000000000000000000000
8306             vec($_, 2, 8) = 2   ==     131072 00000000000000000100000000000000
8307             vec($_, 3, 8) = 2   ==   33554432 00000000000000000000000001000000
8308             vec($_, 0, 8) = 4   ==          4 00100000000000000000000000000000
8309             vec($_, 1, 8) = 4   ==       1024 00000000001000000000000000000000
8310             vec($_, 2, 8) = 4   ==     262144 00000000000000000010000000000000
8311             vec($_, 3, 8) = 4   ==   67108864 00000000000000000000000000100000
8312             vec($_, 0, 8) = 8   ==          8 00010000000000000000000000000000
8313             vec($_, 1, 8) = 8   ==       2048 00000000000100000000000000000000
8314             vec($_, 2, 8) = 8   ==     524288 00000000000000000001000000000000
8315             vec($_, 3, 8) = 8   ==  134217728 00000000000000000000000000010000
8316             vec($_, 0, 8) = 16  ==         16 00001000000000000000000000000000
8317             vec($_, 1, 8) = 16  ==       4096 00000000000010000000000000000000
8318             vec($_, 2, 8) = 16  ==    1048576 00000000000000000000100000000000
8319             vec($_, 3, 8) = 16  ==  268435456 00000000000000000000000000001000
8320             vec($_, 0, 8) = 32  ==         32 00000100000000000000000000000000
8321             vec($_, 1, 8) = 32  ==       8192 00000000000001000000000000000000
8322             vec($_, 2, 8) = 32  ==    2097152 00000000000000000000010000000000
8323             vec($_, 3, 8) = 32  ==  536870912 00000000000000000000000000000100
8324             vec($_, 0, 8) = 64  ==         64 00000010000000000000000000000000
8325             vec($_, 1, 8) = 64  ==      16384 00000000000000100000000000000000
8326             vec($_, 2, 8) = 64  ==    4194304 00000000000000000000001000000000
8327             vec($_, 3, 8) = 64  == 1073741824 00000000000000000000000000000010
8328             vec($_, 0, 8) = 128 ==        128 00000001000000000000000000000000
8329             vec($_, 1, 8) = 128 ==      32768 00000000000000010000000000000000
8330             vec($_, 2, 8) = 128 ==    8388608 00000000000000000000000100000000
8331             vec($_, 3, 8) = 128 == 2147483648 00000000000000000000000000000001
8332
8333       wait
8334           Behaves like wait(2) on your system: it waits for a child process
8335           to terminate and returns the pid of the deceased process, or "-1"
8336           if there are no child processes.  The status is returned in $? and
8337           "${^CHILD_ERROR_NATIVE}".  Note that a return value of "-1" could
8338           mean that child processes are being automatically reaped, as
8339           described in perlipc.
8340
8341           If you use "wait" in your handler for $SIG{CHLD}, it may
8342           accidentally wait for the child created by "qx" or "system".  See
8343           perlipc for details.
8344
8345           Portability issues: "wait" in perlport.
8346
8347       waitpid PID,FLAGS
8348           Waits for a particular child process to terminate and returns the
8349           pid of the deceased process, or "-1" if there is no such child
8350           process.  A non-blocking wait (with WNOHANG in FLAGS) can return 0
8351           if there are child processes matching PID but none have terminated
8352           yet.  The status is returned in $? and "${^CHILD_ERROR_NATIVE}".
8353
8354           A PID of 0 indicates to wait for any child process whose process
8355           group ID is equal to that of the current process.  A PID of less
8356           than "-1" indicates to wait for any child process whose process
8357           group ID is equal to -PID.  A PID of "-1" indicates to wait for any
8358           child process.
8359
8360           If you say
8361
8362               use POSIX ":sys_wait_h";
8363
8364               my $kid;
8365               do {
8366                   $kid = waitpid(-1, WNOHANG);
8367               } while $kid > 0;
8368
8369           or
8370
8371               1 while waitpid(-1, WNOHANG) > 0;
8372
8373           then you can do a non-blocking wait for all pending zombie
8374           processes (see "WAIT" in POSIX).  Non-blocking wait is available on
8375           machines supporting either the waitpid(2) or wait4(2) syscalls.
8376           However, waiting for a particular pid with FLAGS of 0 is
8377           implemented everywhere.  (Perl emulates the system call by
8378           remembering the status values of processes that have exited but
8379           have not been harvested by the Perl script yet.)
8380
8381           Note that on some systems, a return value of "-1" could mean that
8382           child processes are being automatically reaped.  See perlipc for
8383           details, and for other examples.
8384
8385           Portability issues: "waitpid" in perlport.
8386
8387       wantarray
8388           Returns true if the context of the currently executing subroutine
8389           or "eval" is looking for a list value.  Returns false if the
8390           context is looking for a scalar.  Returns the undefined value if
8391           the context is looking for no value (void context).
8392
8393               return unless defined wantarray; # don't bother doing more
8394               my @a = complex_calculation();
8395               return wantarray ? @a : "@a";
8396
8397           "wantarray"'s result is unspecified in the top level of a file, in
8398           a "BEGIN", "UNITCHECK", "CHECK", "INIT" or "END" block, or in a
8399           "DESTROY" method.
8400
8401           This function should have been named wantlist() instead.
8402
8403       warn LIST
8404           Emits a warning, usually by printing it to "STDERR".  "warn"
8405           interprets its operand LIST in the same way as "die", but is
8406           slightly different in what it defaults to when LIST is empty or
8407           makes an empty string.  If it is empty and $@ already contains an
8408           exception value then that value is used after appending
8409           "\t...caught".  If it is empty and $@ is also empty then the string
8410           "Warning: Something's wrong" is used.
8411
8412           By default, the exception derived from the operand LIST is
8413           stringified and printed to "STDERR".  This behaviour can be altered
8414           by installing a $SIG{__WARN__} handler.  If there is such a handler
8415           then no message is automatically printed; it is the handler's
8416           responsibility to deal with the exception as it sees fit (like, for
8417           instance, converting it into a "die").  Most handlers must
8418           therefore arrange to actually display the warnings that they are
8419           not prepared to deal with, by calling "warn" again in the handler.
8420           Note that this is quite safe and will not produce an endless loop,
8421           since "__WARN__" hooks are not called from inside one.
8422
8423           You will find this behavior is slightly different from that of
8424           $SIG{__DIE__} handlers (which don't suppress the error text, but
8425           can instead call "die" again to change it).
8426
8427           Using a "__WARN__" handler provides a powerful way to silence all
8428           warnings (even the so-called mandatory ones).  An example:
8429
8430               # wipe out *all* compile-time warnings
8431               BEGIN { $SIG{'__WARN__'} = sub { warn $_[0] if $DOWARN } }
8432               my $foo = 10;
8433               my $foo = 20;          # no warning about duplicate my $foo,
8434                                      # but hey, you asked for it!
8435               # no compile-time or run-time warnings before here
8436               $DOWARN = 1;
8437
8438               # run-time warnings enabled after here
8439               warn "\$foo is alive and $foo!";     # does show up
8440
8441           See perlvar for details on setting %SIG entries and for more
8442           examples.  See the Carp module for other kinds of warnings using
8443           its "carp" and "cluck" functions.
8444
8445       write FILEHANDLE
8446       write EXPR
8447       write
8448           Writes a formatted record (possibly multi-line) to the specified
8449           FILEHANDLE, using the format associated with that file.  By default
8450           the format for a file is the one having the same name as the
8451           filehandle, but the format for the current output channel (see the
8452           "select" function) may be set explicitly by assigning the name of
8453           the format to the $~ variable.
8454
8455           Top of form processing is handled automatically:  if there is
8456           insufficient room on the current page for the formatted record, the
8457           page is advanced by writing a form feed and a special top-of-page
8458           format is used to format the new page header before the record is
8459           written.  By default, the top-of-page format is the name of the
8460           filehandle with "_TOP" appended, or "top" in the current package if
8461           the former does not exist.  This would be a problem with
8462           autovivified filehandles, but it may be dynamically set to the
8463           format of your choice by assigning the name to the $^ variable
8464           while that filehandle is selected.  The number of lines remaining
8465           on the current page is in variable "$-", which can be set to 0 to
8466           force a new page.
8467
8468           If FILEHANDLE is unspecified, output goes to the current default
8469           output channel, which starts out as STDOUT but may be changed by
8470           the "select" operator.  If the FILEHANDLE is an EXPR, then the
8471           expression is evaluated and the resulting string is used to look up
8472           the name of the FILEHANDLE at run time.  For more on formats, see
8473           perlform.
8474
8475           Note that write is not the opposite of "read".  Unfortunately.
8476
8477       y///
8478           The transliteration operator.  Same as "tr///".  See "Quote-Like
8479           Operators" in perlop.
8480
8481   Non-function Keywords by Cross-reference
8482       perldata
8483
8484       __DATA__
8485       __END__
8486           These keywords are documented in "Special Literals" in perldata.
8487
8488       perlmod
8489
8490       BEGIN
8491       CHECK
8492       END
8493       INIT
8494       UNITCHECK
8495           These compile phase keywords are documented in "BEGIN, UNITCHECK,
8496           CHECK, INIT and END" in perlmod.
8497
8498       perlobj
8499
8500       DESTROY
8501           This method keyword is documented in "Destructors" in perlobj.
8502
8503       perlop
8504
8505       and
8506       cmp
8507       eq
8508       ge
8509       gt
8510       le
8511       lt
8512       ne
8513       not
8514       or
8515       x
8516       xor These operators are documented in perlop.
8517
8518       perlsub
8519
8520       AUTOLOAD
8521           This keyword is documented in "Autoloading" in perlsub.
8522
8523       perlsyn
8524
8525       else
8526       elsif
8527       for
8528       foreach
8529       if
8530       unless
8531       until
8532       while
8533           These flow-control keywords are documented in "Compound Statements"
8534           in perlsyn.
8535
8536       elseif
8537           The "else if" keyword is spelled "elsif" in Perl.  There's no
8538           "elif" or "else if" either.  It does parse "elseif", but only to
8539           warn you about not using it.
8540
8541           See the documentation for flow-control keywords in "Compound
8542           Statements" in perlsyn.
8543
8544       default
8545       given
8546       when
8547           These flow-control keywords related to the experimental switch
8548           feature are documented in "Switch Statements" in perlsyn.
8549
8550
8551
8552perl v5.32.1                      2021-05-31                       PERLFUNC(1)
Impressum