xref: /openbsd/gnu/usr.bin/perl/pod/perlfunc.pod (revision 3bef86f7)
1=head1 NAME
2X<function>
3
4perlfunc - Perl builtin functions
5
6=head1 DESCRIPTION
7
8The functions in this section can serve as terms in an expression.
9They fall into two major categories: list operators and named unary
10operators.  These differ in their precedence relationship with a
11following comma.  (See the precedence table in L<perlop>.)  List
12operators take more than one argument, while unary operators can never
13take more than one argument.  Thus, a comma terminates the argument of
14a unary operator, but merely separates the arguments of a list
15operator.  A unary operator generally provides scalar context to its
16argument, while a list operator may provide either scalar or list
17contexts for its arguments.  If it does both, scalar arguments
18come first and list argument follow, and there can only ever
19be one such list argument.  For instance,
20L<C<splice>|/splice ARRAY,OFFSET,LENGTH,LIST> has three scalar arguments
21followed by a list, whereas L<C<gethostbyname>|/gethostbyname NAME> has
22four scalar arguments.
23
24In the syntax descriptions that follow, list operators that expect a
25list (and provide list context for elements of the list) are shown
26with LIST as an argument.  Such a list may consist of any combination
27of scalar arguments or list values; the list values will be included
28in the list as if each individual element were interpolated at that
29point in the list, forming a longer single-dimensional list value.
30Commas should separate literal elements of the LIST.
31
32Any function in the list below may be used either with or without
33parentheses around its arguments.  (The syntax descriptions omit the
34parentheses.)  If you use parentheses, the simple but occasionally
35surprising rule is this: It I<looks> like a function, therefore it I<is> a
36function, and precedence doesn't matter.  Otherwise it's a list
37operator or unary operator, and precedence does matter.  Whitespace
38between the function and left parenthesis doesn't count, so sometimes
39you need to be careful:
40
41    print 1+2+4;      # Prints 7.
42    print(1+2) + 4;   # Prints 3.
43    print (1+2)+4;    # Also prints 3!
44    print +(1+2)+4;   # Prints 7.
45    print ((1+2)+4);  # Prints 7.
46
47If you run Perl with the L<C<use warnings>|warnings> pragma, it can warn
48you about this.  For example, the third line above produces:
49
50    print (...) interpreted as function at - line 1.
51    Useless use of integer addition in void context at - line 1.
52
53A few functions take no arguments at all, and therefore work as neither
54unary nor list operators.  These include such functions as
55L<C<time>|/time> and L<C<endpwent>|/endpwent>.  For example,
56C<time+86_400> always means C<time() + 86_400>.
57
58For functions that can be used in either a scalar or list context,
59nonabortive failure is generally indicated in scalar context by
60returning the undefined value, and in list context by returning the
61empty list.
62
63Remember the following important rule: There is B<no rule> that relates
64the behavior of an expression in list context to its behavior in scalar
65context, or vice versa.  It might do two totally different things.
66Each operator and function decides which sort of value would be most
67appropriate to return in scalar context.  Some operators return the
68length of the list that would have been returned in list context.  Some
69operators return the first value in the list.  Some operators return the
70last value in the list.  Some operators return a count of successful
71operations.  In general, they do what you want, unless you want
72consistency.
73X<context>
74
75A named array in scalar context is quite different from what would at
76first glance appear to be a list in scalar context.  You can't get a list
77like C<(1,2,3)> into being in scalar context, because the compiler knows
78the context at compile time.  It would generate the scalar comma operator
79there, not the list concatenation version of the comma.  That means it
80was never a list to start with.
81
82In general, functions in Perl that serve as wrappers for system calls
83("syscalls") of the same name (like L<chown(2)>, L<fork(2)>,
84L<closedir(2)>, etc.) return true when they succeed and
85L<C<undef>|/undef EXPR> otherwise, as is usually mentioned in the
86descriptions below.  This is different from the C interfaces, which
87return C<-1> on failure.  Exceptions to this rule include
88L<C<wait>|/wait>, L<C<waitpid>|/waitpid PID,FLAGS>, and
89L<C<syscall>|/syscall NUMBER, LIST>.  System calls also set the special
90L<C<$!>|perlvar/$!> variable on failure.  Other functions do not, except
91accidentally.
92
93Extension modules can also hook into the Perl parser to define new
94kinds of keyword-headed expression.  These may look like functions, but
95may also look completely different.  The syntax following the keyword
96is defined entirely by the extension.  If you are an implementor, see
97L<perlapi/PL_keyword_plugin> for the mechanism.  If you are using such
98a module, see the module's documentation for details of the syntax that
99it defines.
100
101=head2 Perl Functions by Category
102X<function>
103
104Here are Perl's functions (including things that look like
105functions, like some keywords and named operators)
106arranged by category.  Some functions appear in more
107than one place.  Any warnings, including those produced by
108keywords, are described in L<perldiag> and L<warnings>.
109
110=over 4
111
112=item Functions for SCALARs or strings
113X<scalar> X<string> X<character>
114
115=for Pod::Functions =String
116
117L<C<chomp>|/chomp VARIABLE>, L<C<chop>|/chop VARIABLE>,
118L<C<chr>|/chr NUMBER>, L<C<crypt>|/crypt PLAINTEXT,SALT>,
119L<C<fc>|/fc EXPR>, L<C<hex>|/hex EXPR>,
120L<C<index>|/index STR,SUBSTR,POSITION>, L<C<lc>|/lc EXPR>,
121L<C<lcfirst>|/lcfirst EXPR>, L<C<length>|/length EXPR>,
122L<C<oct>|/oct EXPR>, L<C<ord>|/ord EXPR>,
123L<C<pack>|/pack TEMPLATE,LIST>,
124L<C<qE<sol>E<sol>>|/qE<sol>STRINGE<sol>>,
125L<C<qqE<sol>E<sol>>|/qqE<sol>STRINGE<sol>>, L<C<reverse>|/reverse LIST>,
126L<C<rindex>|/rindex STR,SUBSTR,POSITION>,
127L<C<sprintf>|/sprintf FORMAT, LIST>,
128L<C<substr>|/substr EXPR,OFFSET,LENGTH,REPLACEMENT>,
129L<C<trE<sol>E<sol>E<sol>>|/trE<sol>E<sol>E<sol>>, L<C<uc>|/uc EXPR>,
130L<C<ucfirst>|/ucfirst EXPR>,
131L<C<yE<sol>E<sol>E<sol>>|/yE<sol>E<sol>E<sol>>
132
133L<C<fc>|/fc EXPR> is available only if the
134L<C<"fc"> feature|feature/The 'fc' feature> is enabled or if it is
135prefixed with C<CORE::>.  The
136L<C<"fc"> feature|feature/The 'fc' feature> is enabled automatically
137with a C<use v5.16> (or higher) declaration in the current scope.
138
139=item Regular expressions and pattern matching
140X<regular expression> X<regex> X<regexp>
141
142=for Pod::Functions =Regexp
143
144L<C<mE<sol>E<sol>>|/mE<sol>E<sol>>, L<C<pos>|/pos SCALAR>,
145L<C<qrE<sol>E<sol>>|/qrE<sol>STRINGE<sol>>,
146L<C<quotemeta>|/quotemeta EXPR>,
147L<C<sE<sol>E<sol>E<sol>>|/sE<sol>E<sol>E<sol>>,
148L<C<split>|/split E<sol>PATTERNE<sol>,EXPR,LIMIT>,
149L<C<study>|/study SCALAR>
150
151=item Numeric functions
152X<numeric> X<number> X<trigonometric> X<trigonometry>
153
154=for Pod::Functions =Math
155
156L<C<abs>|/abs VALUE>, L<C<atan2>|/atan2 Y,X>, L<C<cos>|/cos EXPR>,
157L<C<exp>|/exp EXPR>, L<C<hex>|/hex EXPR>, L<C<int>|/int EXPR>,
158L<C<log>|/log EXPR>, L<C<oct>|/oct EXPR>, L<C<rand>|/rand EXPR>,
159L<C<sin>|/sin EXPR>, L<C<sqrt>|/sqrt EXPR>, L<C<srand>|/srand EXPR>
160
161=item Functions for real @ARRAYs
162X<array>
163
164=for Pod::Functions =ARRAY
165
166L<C<each>|/each HASH>, L<C<keys>|/keys HASH>, L<C<pop>|/pop ARRAY>,
167L<C<push>|/push ARRAY,LIST>, L<C<shift>|/shift ARRAY>,
168L<C<splice>|/splice ARRAY,OFFSET,LENGTH,LIST>,
169L<C<unshift>|/unshift ARRAY,LIST>, L<C<values>|/values HASH>
170
171=item Functions for list data
172X<list>
173
174=for Pod::Functions =LIST
175
176L<C<grep>|/grep BLOCK LIST>, L<C<join>|/join EXPR,LIST>,
177L<C<map>|/map BLOCK LIST>, L<C<qwE<sol>E<sol>>|/qwE<sol>STRINGE<sol>>,
178L<C<reverse>|/reverse LIST>, L<C<sort>|/sort SUBNAME LIST>,
179L<C<unpack>|/unpack TEMPLATE,EXPR>
180
181=item Functions for real %HASHes
182X<hash>
183
184=for Pod::Functions =HASH
185
186L<C<delete>|/delete EXPR>, L<C<each>|/each HASH>,
187L<C<exists>|/exists EXPR>, L<C<keys>|/keys HASH>,
188L<C<values>|/values HASH>
189
190=item Input and output functions
191X<I/O> X<input> X<output> X<dbm>
192
193=for Pod::Functions =I/O
194
195L<C<binmode>|/binmode FILEHANDLE, LAYER>, L<C<close>|/close FILEHANDLE>,
196L<C<closedir>|/closedir DIRHANDLE>, L<C<dbmclose>|/dbmclose HASH>,
197L<C<dbmopen>|/dbmopen HASH,DBNAME,MASK>, L<C<die>|/die LIST>,
198L<C<eof>|/eof FILEHANDLE>, L<C<fileno>|/fileno FILEHANDLE>,
199L<C<flock>|/flock FILEHANDLE,OPERATION>, L<C<format>|/format>,
200L<C<getc>|/getc FILEHANDLE>, L<C<print>|/print FILEHANDLE LIST>,
201L<C<printf>|/printf FILEHANDLE FORMAT, LIST>,
202L<C<read>|/read FILEHANDLE,SCALAR,LENGTH,OFFSET>,
203L<C<readdir>|/readdir DIRHANDLE>, L<C<readline>|/readline EXPR>,
204L<C<rewinddir>|/rewinddir DIRHANDLE>, L<C<say>|/say FILEHANDLE LIST>,
205L<C<seek>|/seek FILEHANDLE,POSITION,WHENCE>,
206L<C<seekdir>|/seekdir DIRHANDLE,POS>,
207L<C<select>|/select RBITS,WBITS,EBITS,TIMEOUT>,
208L<C<syscall>|/syscall NUMBER, LIST>,
209L<C<sysread>|/sysread FILEHANDLE,SCALAR,LENGTH,OFFSET>,
210L<C<sysseek>|/sysseek FILEHANDLE,POSITION,WHENCE>,
211L<C<syswrite>|/syswrite FILEHANDLE,SCALAR,LENGTH,OFFSET>,
212L<C<tell>|/tell FILEHANDLE>, L<C<telldir>|/telldir DIRHANDLE>,
213L<C<truncate>|/truncate FILEHANDLE,LENGTH>, L<C<warn>|/warn LIST>,
214L<C<write>|/write FILEHANDLE>
215
216L<C<say>|/say FILEHANDLE LIST> is available only if the
217L<C<"say"> feature|feature/The 'say' feature> is enabled or if it is
218prefixed with C<CORE::>.  The
219L<C<"say"> feature|feature/The 'say' feature> is enabled automatically
220with a C<use v5.10> (or higher) declaration in the current scope.
221
222=item Functions for fixed-length data or records
223
224=for Pod::Functions =Binary
225
226L<C<pack>|/pack TEMPLATE,LIST>,
227L<C<read>|/read FILEHANDLE,SCALAR,LENGTH,OFFSET>,
228L<C<syscall>|/syscall NUMBER, LIST>,
229L<C<sysread>|/sysread FILEHANDLE,SCALAR,LENGTH,OFFSET>,
230L<C<sysseek>|/sysseek FILEHANDLE,POSITION,WHENCE>,
231L<C<syswrite>|/syswrite FILEHANDLE,SCALAR,LENGTH,OFFSET>,
232L<C<unpack>|/unpack TEMPLATE,EXPR>, L<C<vec>|/vec EXPR,OFFSET,BITS>
233
234=item Functions for filehandles, files, or directories
235X<file> X<filehandle> X<directory> X<pipe> X<link> X<symlink>
236
237=for Pod::Functions =File
238
239L<C<-I<X>>|/-X FILEHANDLE>, L<C<chdir>|/chdir EXPR>,
240L<C<chmod>|/chmod LIST>, L<C<chown>|/chown LIST>,
241L<C<chroot>|/chroot FILENAME>,
242L<C<fcntl>|/fcntl FILEHANDLE,FUNCTION,SCALAR>, L<C<glob>|/glob EXPR>,
243L<C<ioctl>|/ioctl FILEHANDLE,FUNCTION,SCALAR>,
244L<C<link>|/link OLDFILE,NEWFILE>, L<C<lstat>|/lstat FILEHANDLE>,
245L<C<mkdir>|/mkdir FILENAME,MODE>, L<C<open>|/open FILEHANDLE,MODE,EXPR>,
246L<C<opendir>|/opendir DIRHANDLE,EXPR>, L<C<readlink>|/readlink EXPR>,
247L<C<rename>|/rename OLDNAME,NEWNAME>, L<C<rmdir>|/rmdir FILENAME>,
248L<C<select>|/select FILEHANDLE>, L<C<stat>|/stat FILEHANDLE>,
249L<C<symlink>|/symlink OLDFILE,NEWFILE>,
250L<C<sysopen>|/sysopen FILEHANDLE,FILENAME,MODE>,
251L<C<umask>|/umask EXPR>, L<C<unlink>|/unlink LIST>,
252L<C<utime>|/utime LIST>
253
254=item Keywords related to the control flow of your Perl program
255X<control flow>
256
257=for Pod::Functions =Flow
258
259L<C<break>|/break>, L<C<caller>|/caller EXPR>,
260L<C<continue>|/continue BLOCK>, L<C<die>|/die LIST>, L<C<do>|/do BLOCK>,
261L<C<dump>|/dump LABEL>, L<C<eval>|/eval EXPR>,
262L<C<evalbytes>|/evalbytes EXPR>, L<C<exit>|/exit EXPR>,
263L<C<__FILE__>|/__FILE__>, L<C<goto>|/goto LABEL>,
264L<C<last>|/last LABEL>, L<C<__LINE__>|/__LINE__>,
265L<C<next>|/next LABEL>, L<C<__PACKAGE__>|/__PACKAGE__>,
266L<C<redo>|/redo LABEL>, L<C<return>|/return EXPR>,
267L<C<sub>|/sub NAME BLOCK>, L<C<__SUB__>|/__SUB__>,
268L<C<wantarray>|/wantarray>
269
270L<C<break>|/break> is available only if you enable the experimental
271L<C<"switch"> feature|feature/The 'switch' feature> or use the C<CORE::>
272prefix.  The L<C<"switch"> feature|feature/The 'switch' feature> also
273enables the C<default>, C<given> and C<when> statements, which are
274documented in L<perlsyn/"Switch Statements">.
275The L<C<"switch"> feature|feature/The 'switch' feature> is enabled
276automatically with a C<use v5.10> (or higher) declaration in the current
277scope.  In Perl v5.14 and earlier, L<C<continue>|/continue BLOCK>
278required the L<C<"switch"> feature|feature/The 'switch' feature>, like
279the other keywords.
280
281L<C<evalbytes>|/evalbytes EXPR> is only available with the
282L<C<"evalbytes"> feature|feature/The 'unicode_eval' and 'evalbytes' features>
283(see L<feature>) or if prefixed with C<CORE::>.  L<C<__SUB__>|/__SUB__>
284is only available with the
285L<C<"current_sub"> feature|feature/The 'current_sub' feature> or if
286prefixed with C<CORE::>.  Both the
287L<C<"evalbytes">|feature/The 'unicode_eval' and 'evalbytes' features>
288and L<C<"current_sub">|feature/The 'current_sub' feature> features are
289enabled automatically with a C<use v5.16> (or higher) declaration in the
290current scope.
291
292=item Keywords related to scoping
293
294=for Pod::Functions =Namespace
295
296L<C<caller>|/caller EXPR>, L<C<import>|/import LIST>,
297L<C<local>|/local EXPR>, L<C<my>|/my VARLIST>, L<C<our>|/our VARLIST>,
298L<C<package>|/package NAMESPACE>, L<C<state>|/state VARLIST>,
299L<C<use>|/use Module VERSION LIST>
300
301L<C<state>|/state VARLIST> is available only if the
302L<C<"state"> feature|feature/The 'state' feature> is enabled or if it is
303prefixed with C<CORE::>.  The
304L<C<"state"> feature|feature/The 'state' feature> is enabled
305automatically with a C<use v5.10> (or higher) declaration in the current
306scope.
307
308=item Miscellaneous functions
309
310=for Pod::Functions =Misc
311
312L<C<defined>|/defined EXPR>, L<C<formline>|/formline PICTURE,LIST>,
313L<C<lock>|/lock THING>, L<C<prototype>|/prototype FUNCTION>,
314L<C<reset>|/reset EXPR>, L<C<scalar>|/scalar EXPR>,
315L<C<undef>|/undef EXPR>
316
317=item Functions for processes and process groups
318X<process> X<pid> X<process id>
319
320=for Pod::Functions =Process
321
322L<C<alarm>|/alarm SECONDS>, L<C<exec>|/exec LIST>, L<C<fork>|/fork>,
323L<C<getpgrp>|/getpgrp PID>, L<C<getppid>|/getppid>,
324L<C<getpriority>|/getpriority WHICH,WHO>, L<C<kill>|/kill SIGNAL, LIST>,
325L<C<pipe>|/pipe READHANDLE,WRITEHANDLE>,
326L<C<qxE<sol>E<sol>>|/qxE<sol>STRINGE<sol>>,
327L<C<readpipe>|/readpipe EXPR>, L<C<setpgrp>|/setpgrp PID,PGRP>,
328L<C<setpriority>|/setpriority WHICH,WHO,PRIORITY>,
329L<C<sleep>|/sleep EXPR>, L<C<system>|/system LIST>, L<C<times>|/times>,
330L<C<wait>|/wait>, L<C<waitpid>|/waitpid PID,FLAGS>
331
332=item Keywords related to Perl modules
333X<module>
334
335=for Pod::Functions =Modules
336
337L<C<do>|/do EXPR>, L<C<import>|/import LIST>,
338L<C<no>|/no MODULE VERSION LIST>, L<C<package>|/package NAMESPACE>,
339L<C<require>|/require VERSION>, L<C<use>|/use Module VERSION LIST>
340
341=item Keywords related to classes and object-orientation
342X<object> X<class> X<package>
343
344=for Pod::Functions =Objects
345
346L<C<bless>|/bless REF,CLASSNAME>, L<C<dbmclose>|/dbmclose HASH>,
347L<C<dbmopen>|/dbmopen HASH,DBNAME,MASK>,
348L<C<package>|/package NAMESPACE>, L<C<ref>|/ref EXPR>,
349L<C<tie>|/tie VARIABLE,CLASSNAME,LIST>, L<C<tied>|/tied VARIABLE>,
350L<C<untie>|/untie VARIABLE>, L<C<use>|/use Module VERSION LIST>
351
352=item Low-level socket functions
353X<socket> X<sock>
354
355=for Pod::Functions =Socket
356
357L<C<accept>|/accept NEWSOCKET,GENERICSOCKET>,
358L<C<bind>|/bind SOCKET,NAME>, L<C<connect>|/connect SOCKET,NAME>,
359L<C<getpeername>|/getpeername SOCKET>,
360L<C<getsockname>|/getsockname SOCKET>,
361L<C<getsockopt>|/getsockopt SOCKET,LEVEL,OPTNAME>,
362L<C<listen>|/listen SOCKET,QUEUESIZE>,
363L<C<recv>|/recv SOCKET,SCALAR,LENGTH,FLAGS>,
364L<C<send>|/send SOCKET,MSG,FLAGS,TO>,
365L<C<setsockopt>|/setsockopt SOCKET,LEVEL,OPTNAME,OPTVAL>,
366L<C<shutdown>|/shutdown SOCKET,HOW>,
367L<C<socket>|/socket SOCKET,DOMAIN,TYPE,PROTOCOL>,
368L<C<socketpair>|/socketpair SOCKET1,SOCKET2,DOMAIN,TYPE,PROTOCOL>
369
370=item System V interprocess communication functions
371X<IPC> X<System V> X<semaphore> X<shared memory> X<memory> X<message>
372
373=for Pod::Functions =SysV
374
375L<C<msgctl>|/msgctl ID,CMD,ARG>, L<C<msgget>|/msgget KEY,FLAGS>,
376L<C<msgrcv>|/msgrcv ID,VAR,SIZE,TYPE,FLAGS>,
377L<C<msgsnd>|/msgsnd ID,MSG,FLAGS>,
378L<C<semctl>|/semctl ID,SEMNUM,CMD,ARG>,
379L<C<semget>|/semget KEY,NSEMS,FLAGS>, L<C<semop>|/semop KEY,OPSTRING>,
380L<C<shmctl>|/shmctl ID,CMD,ARG>, L<C<shmget>|/shmget KEY,SIZE,FLAGS>,
381L<C<shmread>|/shmread ID,VAR,POS,SIZE>,
382L<C<shmwrite>|/shmwrite ID,STRING,POS,SIZE>
383
384=item Fetching user and group info
385X<user> X<group> X<password> X<uid> X<gid>  X<passwd> X</etc/passwd>
386
387=for Pod::Functions =User
388
389L<C<endgrent>|/endgrent>, L<C<endhostent>|/endhostent>,
390L<C<endnetent>|/endnetent>, L<C<endpwent>|/endpwent>,
391L<C<getgrent>|/getgrent>, L<C<getgrgid>|/getgrgid GID>,
392L<C<getgrnam>|/getgrnam NAME>, L<C<getlogin>|/getlogin>,
393L<C<getpwent>|/getpwent>, L<C<getpwnam>|/getpwnam NAME>,
394L<C<getpwuid>|/getpwuid UID>, L<C<setgrent>|/setgrent>,
395L<C<setpwent>|/setpwent>
396
397=item Fetching network info
398X<network> X<protocol> X<host> X<hostname> X<IP> X<address> X<service>
399
400=for Pod::Functions =Network
401
402L<C<endprotoent>|/endprotoent>, L<C<endservent>|/endservent>,
403L<C<gethostbyaddr>|/gethostbyaddr ADDR,ADDRTYPE>,
404L<C<gethostbyname>|/gethostbyname NAME>, L<C<gethostent>|/gethostent>,
405L<C<getnetbyaddr>|/getnetbyaddr ADDR,ADDRTYPE>,
406L<C<getnetbyname>|/getnetbyname NAME>, L<C<getnetent>|/getnetent>,
407L<C<getprotobyname>|/getprotobyname NAME>,
408L<C<getprotobynumber>|/getprotobynumber NUMBER>,
409L<C<getprotoent>|/getprotoent>,
410L<C<getservbyname>|/getservbyname NAME,PROTO>,
411L<C<getservbyport>|/getservbyport PORT,PROTO>,
412L<C<getservent>|/getservent>, L<C<sethostent>|/sethostent STAYOPEN>,
413L<C<setnetent>|/setnetent STAYOPEN>,
414L<C<setprotoent>|/setprotoent STAYOPEN>,
415L<C<setservent>|/setservent STAYOPEN>
416
417=item Time-related functions
418X<time> X<date>
419
420=for Pod::Functions =Time
421
422L<C<gmtime>|/gmtime EXPR>, L<C<localtime>|/localtime EXPR>,
423L<C<time>|/time>, L<C<times>|/times>
424
425=item Non-function keywords
426
427=for Pod::Functions =!Non-functions
428
429C<and>,
430C<AUTOLOAD>,
431C<BEGIN>,
432C<catch>,
433C<CHECK>,
434C<cmp>,
435C<CORE>,
436C<__DATA__>,
437C<default>,
438C<defer>,
439C<DESTROY>,
440C<else>,
441C<elseif>,
442C<elsif>,
443C<END>,
444C<__END__>,
445C<eq>,
446C<finally>,
447C<for>,
448C<foreach>,
449C<ge>,
450C<given>,
451C<gt>,
452C<if>,
453C<INIT>,
454C<isa>,
455C<le>,
456C<lt>,
457C<ne>,
458C<not>,
459C<or>,
460C<try>,
461C<UNITCHECK>,
462C<unless>,
463C<until>,
464C<when>,
465C<while>,
466C<x>,
467C<xor>
468
469=back
470
471=head2 Portability
472X<portability> X<Unix> X<portable>
473
474Perl was born in Unix and can therefore access all common Unix
475system calls.  In non-Unix environments, the functionality of some
476Unix system calls may not be available or details of the available
477functionality may differ slightly.  The Perl functions affected
478by this are:
479
480L<C<-I<X>>|/-X FILEHANDLE>, L<C<binmode>|/binmode FILEHANDLE, LAYER>,
481L<C<chmod>|/chmod LIST>, L<C<chown>|/chown LIST>,
482L<C<chroot>|/chroot FILENAME>, L<C<crypt>|/crypt PLAINTEXT,SALT>,
483L<C<dbmclose>|/dbmclose HASH>, L<C<dbmopen>|/dbmopen HASH,DBNAME,MASK>,
484L<C<dump>|/dump LABEL>, L<C<endgrent>|/endgrent>,
485L<C<endhostent>|/endhostent>, L<C<endnetent>|/endnetent>,
486L<C<endprotoent>|/endprotoent>, L<C<endpwent>|/endpwent>,
487L<C<endservent>|/endservent>, L<C<exec>|/exec LIST>,
488L<C<fcntl>|/fcntl FILEHANDLE,FUNCTION,SCALAR>,
489L<C<flock>|/flock FILEHANDLE,OPERATION>, L<C<fork>|/fork>,
490L<C<getgrent>|/getgrent>, L<C<getgrgid>|/getgrgid GID>,
491L<C<gethostbyname>|/gethostbyname NAME>, L<C<gethostent>|/gethostent>,
492L<C<getlogin>|/getlogin>,
493L<C<getnetbyaddr>|/getnetbyaddr ADDR,ADDRTYPE>,
494L<C<getnetbyname>|/getnetbyname NAME>, L<C<getnetent>|/getnetent>,
495L<C<getppid>|/getppid>, L<C<getpgrp>|/getpgrp PID>,
496L<C<getpriority>|/getpriority WHICH,WHO>,
497L<C<getprotobynumber>|/getprotobynumber NUMBER>,
498L<C<getprotoent>|/getprotoent>, L<C<getpwent>|/getpwent>,
499L<C<getpwnam>|/getpwnam NAME>, L<C<getpwuid>|/getpwuid UID>,
500L<C<getservbyport>|/getservbyport PORT,PROTO>,
501L<C<getservent>|/getservent>,
502L<C<getsockopt>|/getsockopt SOCKET,LEVEL,OPTNAME>,
503L<C<glob>|/glob EXPR>, L<C<ioctl>|/ioctl FILEHANDLE,FUNCTION,SCALAR>,
504L<C<kill>|/kill SIGNAL, LIST>, L<C<link>|/link OLDFILE,NEWFILE>,
505L<C<lstat>|/lstat FILEHANDLE>, L<C<msgctl>|/msgctl ID,CMD,ARG>,
506L<C<msgget>|/msgget KEY,FLAGS>,
507L<C<msgrcv>|/msgrcv ID,VAR,SIZE,TYPE,FLAGS>,
508L<C<msgsnd>|/msgsnd ID,MSG,FLAGS>, L<C<open>|/open FILEHANDLE,MODE,EXPR>,
509L<C<pipe>|/pipe READHANDLE,WRITEHANDLE>, L<C<readlink>|/readlink EXPR>,
510L<C<rename>|/rename OLDNAME,NEWNAME>,
511L<C<select>|/select RBITS,WBITS,EBITS,TIMEOUT>,
512L<C<semctl>|/semctl ID,SEMNUM,CMD,ARG>,
513L<C<semget>|/semget KEY,NSEMS,FLAGS>, L<C<semop>|/semop KEY,OPSTRING>,
514L<C<setgrent>|/setgrent>, L<C<sethostent>|/sethostent STAYOPEN>,
515L<C<setnetent>|/setnetent STAYOPEN>, L<C<setpgrp>|/setpgrp PID,PGRP>,
516L<C<setpriority>|/setpriority WHICH,WHO,PRIORITY>,
517L<C<setprotoent>|/setprotoent STAYOPEN>, L<C<setpwent>|/setpwent>,
518L<C<setservent>|/setservent STAYOPEN>,
519L<C<setsockopt>|/setsockopt SOCKET,LEVEL,OPTNAME,OPTVAL>,
520L<C<shmctl>|/shmctl ID,CMD,ARG>, L<C<shmget>|/shmget KEY,SIZE,FLAGS>,
521L<C<shmread>|/shmread ID,VAR,POS,SIZE>,
522L<C<shmwrite>|/shmwrite ID,STRING,POS,SIZE>,
523L<C<socket>|/socket SOCKET,DOMAIN,TYPE,PROTOCOL>,
524L<C<socketpair>|/socketpair SOCKET1,SOCKET2,DOMAIN,TYPE,PROTOCOL>,
525L<C<stat>|/stat FILEHANDLE>, L<C<symlink>|/symlink OLDFILE,NEWFILE>,
526L<C<syscall>|/syscall NUMBER, LIST>,
527L<C<sysopen>|/sysopen FILEHANDLE,FILENAME,MODE>,
528L<C<system>|/system LIST>, L<C<times>|/times>,
529L<C<truncate>|/truncate FILEHANDLE,LENGTH>, L<C<umask>|/umask EXPR>,
530L<C<unlink>|/unlink LIST>, L<C<utime>|/utime LIST>, L<C<wait>|/wait>,
531L<C<waitpid>|/waitpid PID,FLAGS>
532
533For more information about the portability of these functions, see
534L<perlport> and other available platform-specific documentation.
535
536=head2 Alphabetical Listing of Perl Functions
537
538=over
539
540=item -X FILEHANDLE
541X<-r>X<-w>X<-x>X<-o>X<-R>X<-W>X<-X>X<-O>X<-e>X<-z>X<-s>X<-f>X<-d>X<-l>X<-p>
542X<-S>X<-b>X<-c>X<-t>X<-u>X<-g>X<-k>X<-T>X<-B>X<-M>X<-A>X<-C>
543
544=item -X EXPR
545
546=item -X DIRHANDLE
547
548=item -X
549
550=for Pod::Functions a file test (-r, -x, etc)
551
552A file test, where X is one of the letters listed below.  This unary
553operator takes one argument, either a filename, a filehandle, or a dirhandle,
554and tests the associated file to see if something is true about it.  If the
555argument is omitted, tests L<C<$_>|perlvar/$_>, except for C<-t>, which
556tests STDIN.  Unless otherwise documented, it returns C<1> for true and
557C<''> for false.  If the file doesn't exist or can't be examined, it
558returns L<C<undef>|/undef EXPR> and sets L<C<$!>|perlvar/$!> (errno).
559With the exception of the C<-l> test they all follow symbolic links
560because they use C<stat()> and not C<lstat()> (so dangling symlinks can't
561be examined and will therefore report failure).
562
563Despite the funny names, precedence is the same as any other named unary
564operator.  The operator may be any of:
565
566    -r  File is readable by effective uid/gid.
567    -w  File is writable by effective uid/gid.
568    -x  File is executable by effective uid/gid.
569    -o  File is owned by effective uid.
570
571    -R  File is readable by real uid/gid.
572    -W  File is writable by real uid/gid.
573    -X  File is executable by real uid/gid.
574    -O  File is owned by real uid.
575
576    -e  File exists.
577    -z  File has zero size (is empty).
578    -s  File has nonzero size (returns size in bytes).
579
580    -f  File is a plain file.
581    -d  File is a directory.
582    -l  File is a symbolic link (false if symlinks aren't
583        supported by the file system).
584    -p  File is a named pipe (FIFO), or Filehandle is a pipe.
585    -S  File is a socket.
586    -b  File is a block special file.
587    -c  File is a character special file.
588    -t  Filehandle is opened to a tty.
589
590    -u  File has setuid bit set.
591    -g  File has setgid bit set.
592    -k  File has sticky bit set.
593
594    -T  File is an ASCII or UTF-8 text file (heuristic guess).
595    -B  File is a "binary" file (opposite of -T).
596
597    -M  Script start time minus file modification time, in days.
598    -A  Same for access time.
599    -C  Same for inode change time (Unix, may differ for other
600	platforms)
601
602Example:
603
604    while (<>) {
605        chomp;
606        next unless -f $_;  # ignore specials
607        #...
608    }
609
610Note that C<-s/a/b/> does not do a negated substitution.  Saying
611C<-exp($foo)> still works as expected, however: only single letters
612following a minus are interpreted as file tests.
613
614These operators are exempt from the "looks like a function rule" described
615above.  That is, an opening parenthesis after the operator does not affect
616how much of the following code constitutes the argument.  Put the opening
617parentheses before the operator to separate it from code that follows (this
618applies only to operators with higher precedence than unary operators, of
619course):
620
621    -s($file) + 1024   # probably wrong; same as -s($file + 1024)
622    (-s $file) + 1024  # correct
623
624The interpretation of the file permission operators C<-r>, C<-R>,
625C<-w>, C<-W>, C<-x>, and C<-X> is by default based solely on the mode
626of the file and the uids and gids of the user.  There may be other
627reasons you can't actually read, write, or execute the file: for
628example network filesystem access controls, ACLs (access control lists),
629read-only filesystems, and unrecognized executable formats.  Note
630that the use of these six specific operators to verify if some operation
631is possible is usually a mistake, because it may be open to race
632conditions.
633
634Also note that, for the superuser on the local filesystems, the C<-r>,
635C<-R>, C<-w>, and C<-W> tests always return 1, and C<-x> and C<-X> return 1
636if any execute bit is set in the mode.  Scripts run by the superuser
637may thus need to do a L<C<stat>|/stat FILEHANDLE> to determine the
638actual mode of the file, or temporarily set their effective uid to
639something else.
640
641If you are using ACLs, there is a pragma called L<C<filetest>|filetest>
642that may produce more accurate results than the bare
643L<C<stat>|/stat FILEHANDLE> mode bits.
644When under C<use filetest 'access'>, the above-mentioned filetests
645test whether the permission can(not) be granted using the L<access(2)>
646family of system calls.  Also note that the C<-x> and C<-X> tests may
647under this pragma return true even if there are no execute permission
648bits set (nor any extra execute permission ACLs).  This strangeness is
649due to the underlying system calls' definitions.  Note also that, due to
650the implementation of C<use filetest 'access'>, the C<_> special
651filehandle won't cache the results of the file tests when this pragma is
652in effect.  Read the documentation for the L<C<filetest>|filetest>
653pragma for more information.
654
655The C<-T> and C<-B> tests work as follows.  The first block or so of
656the file is examined to see if it is valid UTF-8 that includes non-ASCII
657characters.  If so, it's a C<-T> file.  Otherwise, that same portion of
658the file is examined for odd characters such as strange control codes or
659characters with the high bit set.  If more than a third of the
660characters are strange, it's a C<-B> file; otherwise it's a C<-T> file.
661Also, any file containing a zero byte in the examined portion is
662considered a binary file.  (If executed within the scope of a L<S<use
663locale>|perllocale> which includes C<LC_CTYPE>, odd characters are
664anything that isn't a printable nor space in the current locale.)  If
665C<-T> or C<-B> is used on a filehandle, the current IO buffer is
666examined
667rather than the first block.  Both C<-T> and C<-B> return true on an empty
668file, or a file at EOF when testing a filehandle.  Because you have to
669read a file to do the C<-T> test, on most occasions you want to use a C<-f>
670against the file first, as in C<next unless -f $file && -T $file>.
671
672If any of the file tests (or either the L<C<stat>|/stat FILEHANDLE> or
673L<C<lstat>|/lstat FILEHANDLE> operator) is given the special filehandle
674consisting of a solitary underline, then the stat structure of the
675previous file test (or L<C<stat>|/stat FILEHANDLE> operator) is used,
676saving a system call.  (This doesn't work with C<-t>, and you need to
677remember that L<C<lstat>|/lstat FILEHANDLE> and C<-l> leave values in
678the stat structure for the symbolic link, not the real file.)  (Also, if
679the stat buffer was filled by an L<C<lstat>|/lstat FILEHANDLE> call,
680C<-T> and C<-B> will reset it with the results of C<stat _>).
681Example:
682
683    print "Can do.\n" if -r $a || -w _ || -x _;
684
685    stat($filename);
686    print "Readable\n" if -r _;
687    print "Writable\n" if -w _;
688    print "Executable\n" if -x _;
689    print "Setuid\n" if -u _;
690    print "Setgid\n" if -g _;
691    print "Sticky\n" if -k _;
692    print "Text\n" if -T _;
693    print "Binary\n" if -B _;
694
695As of Perl 5.10.0, as a form of purely syntactic sugar, you can stack file
696test operators, in a way that C<-f -w -x $file> is equivalent to
697C<-x $file && -w _ && -f _>.  (This is only fancy syntax: if you use
698the return value of C<-f $file> as an argument to another filetest
699operator, no special magic will happen.)
700
701Portability issues: L<perlport/-X>.
702
703To avoid confusing would-be users of your code with mysterious
704syntax errors, put something like this at the top of your script:
705
706    use v5.10;  # so filetest ops can stack
707
708=item abs VALUE
709X<abs> X<absolute>
710
711=item abs
712
713=for Pod::Functions absolute value function
714
715Returns the absolute value of its argument.
716If VALUE is omitted, uses L<C<$_>|perlvar/$_>.
717
718=item accept NEWSOCKET,GENERICSOCKET
719X<accept>
720
721=for Pod::Functions accept an incoming socket connect
722
723Accepts an incoming socket connect, just as L<accept(2)>
724does.  Returns the packed address if it succeeded, false otherwise.
725See the example in L<perlipc/"Sockets: Client/Server Communication">.
726
727On systems that support a close-on-exec flag on files, the flag will
728be set for the newly opened file descriptor, as determined by the
729value of L<C<$^F>|perlvar/$^F>.  See L<perlvar/$^F>.
730
731=item alarm SECONDS
732X<alarm>
733X<SIGALRM>
734X<timer>
735
736=item alarm
737
738=for Pod::Functions schedule a SIGALRM
739
740Arranges to have a SIGALRM delivered to this process after the
741specified number of wallclock seconds has elapsed.  If SECONDS is not
742specified, the value stored in L<C<$_>|perlvar/$_> is used.  (On some
743machines, unfortunately, the elapsed time may be up to one second less
744or more than you specified because of how seconds are counted, and
745process scheduling may delay the delivery of the signal even further.)
746
747Only one timer may be counting at once.  Each call disables the
748previous timer, and an argument of C<0> may be supplied to cancel the
749previous timer without starting a new one.  The returned value is the
750amount of time remaining on the previous timer.
751
752For delays of finer granularity than one second, the L<Time::HiRes> module
753(from CPAN, and starting from Perl 5.8 part of the standard
754distribution) provides
755L<C<ualarm>|Time::HiRes/ualarm ( $useconds [, $interval_useconds ] )>.
756You may also use Perl's four-argument version of
757L<C<select>|/select RBITS,WBITS,EBITS,TIMEOUT> leaving the first three
758arguments undefined, or you might be able to use the
759L<C<syscall>|/syscall NUMBER, LIST> interface to access L<setitimer(2)>
760if your system supports it.  See L<perlfaq8> for details.
761
762It is usually a mistake to intermix L<C<alarm>|/alarm SECONDS> and
763L<C<sleep>|/sleep EXPR> calls, because L<C<sleep>|/sleep EXPR> may be
764internally implemented on your system with L<C<alarm>|/alarm SECONDS>.
765
766If you want to use L<C<alarm>|/alarm SECONDS> to time out a system call
767you need to use an L<C<eval>|/eval EXPR>/L<C<die>|/die LIST> pair.  You
768can't rely on the alarm causing the system call to fail with
769L<C<$!>|perlvar/$!> set to C<EINTR> because Perl sets up signal handlers
770to restart system calls on some systems.  Using
771L<C<eval>|/eval EXPR>/L<C<die>|/die LIST> always works, modulo the
772caveats given in L<perlipc/"Signals">.
773
774    eval {
775        local $SIG{ALRM} = sub { die "alarm\n" }; # NB: \n required
776        alarm $timeout;
777        my $nread = sysread $socket, $buffer, $size;
778        alarm 0;
779    };
780    if ($@) {
781        die unless $@ eq "alarm\n";   # propagate unexpected errors
782        # timed out
783    }
784    else {
785        # didn't
786    }
787
788For more information see L<perlipc>.
789
790Portability issues: L<perlport/alarm>.
791
792=item atan2 Y,X
793X<atan2> X<arctangent> X<tan> X<tangent>
794
795=for Pod::Functions arctangent of Y/X in the range -PI to PI
796
797Returns the arctangent of Y/X in the range -PI to PI.
798
799For the tangent operation, you may use the
800L<C<Math::Trig::tan>|Math::Trig/B<tan>> function, or use the familiar
801relation:
802
803    sub tan { sin($_[0]) / cos($_[0])  }
804
805The return value for C<atan2(0,0)> is implementation-defined; consult
806your L<atan2(3)> manpage for more information.
807
808Portability issues: L<perlport/atan2>.
809
810=item bind SOCKET,NAME
811X<bind>
812
813=for Pod::Functions binds an address to a socket
814
815Binds a network address to a socket, just as L<bind(2)>
816does.  Returns true if it succeeded, false otherwise.  NAME should be a
817packed address of the appropriate type for the socket.  See the examples in
818L<perlipc/"Sockets: Client/Server Communication">.
819
820=item binmode FILEHANDLE, LAYER
821X<binmode> X<binary> X<text> X<DOS> X<Windows>
822
823=item binmode FILEHANDLE
824
825=for Pod::Functions prepare binary files for I/O
826
827Arranges for FILEHANDLE to be read or written in "binary" or "text"
828mode on systems where the run-time libraries distinguish between
829binary and text files.  If FILEHANDLE is an expression, the value is
830taken as the name of the filehandle.  Returns true on success,
831otherwise it returns L<C<undef>|/undef EXPR> and sets
832L<C<$!>|perlvar/$!> (errno).
833
834On some systems (in general, DOS- and Windows-based systems)
835L<C<binmode>|/binmode FILEHANDLE, LAYER> is necessary when you're not
836working with a text file.  For the sake of portability it is a good idea
837always to use it when appropriate, and never to use it when it isn't
838appropriate.  Also, people can set their I/O to be by default
839UTF8-encoded Unicode, not bytes.
840
841In other words: regardless of platform, use
842L<C<binmode>|/binmode FILEHANDLE, LAYER> on binary data, like images,
843for example.
844
845If LAYER is present it is a single string, but may contain multiple
846directives.  The directives alter the behaviour of the filehandle.
847When LAYER is present, using binmode on a text file makes sense.
848
849If LAYER is omitted or specified as C<:raw> the filehandle is made
850suitable for passing binary data.  This includes turning off possible CRLF
851translation and marking it as bytes (as opposed to Unicode characters).
852Note that, despite what may be implied in I<"Programming Perl"> (the
853Camel, 3rd edition) or elsewhere, C<:raw> is I<not> simply the inverse of C<:crlf>.
854Other layers that would affect the binary nature of the stream are
855I<also> disabled.  See L<PerlIO>, and the discussion about the PERLIO
856environment variable in L<perlrun|perlrun/PERLIO>.
857
858The C<:bytes>, C<:crlf>, C<:utf8>, and any other directives of the
859form C<:...>, are called I/O I<layers>.  The L<open> pragma can be used to
860establish default I/O layers.
861
862I<The LAYER parameter of the L<C<binmode>|/binmode FILEHANDLE, LAYER>
863function is described as "DISCIPLINE" in "Programming Perl, 3rd
864Edition".  However, since the publishing of this book, by many known as
865"Camel III", the consensus of the naming of this functionality has moved
866from "discipline" to "layer".  All documentation of this version of Perl
867therefore refers to "layers" rather than to "disciplines".  Now back to
868the regularly scheduled documentation...>
869
870To mark FILEHANDLE as UTF-8, use C<:utf8> or C<:encoding(UTF-8)>.
871C<:utf8> just marks the data as UTF-8 without further checking,
872while C<:encoding(UTF-8)> checks the data for actually being valid
873UTF-8.  More details can be found in L<PerlIO::encoding>.
874
875In general, L<C<binmode>|/binmode FILEHANDLE, LAYER> should be called
876after L<C<open>|/open FILEHANDLE,MODE,EXPR> but before any I/O is done on the
877filehandle.  Calling L<C<binmode>|/binmode FILEHANDLE, LAYER> normally
878flushes any pending buffered output data (and perhaps pending input
879data) on the handle.  An exception to this is the C<:encoding> layer
880that changes the default character encoding of the handle.
881The C<:encoding> layer sometimes needs to be called in
882mid-stream, and it doesn't flush the stream.  C<:encoding>
883also implicitly pushes on top of itself the C<:utf8> layer because
884internally Perl operates on UTF8-encoded Unicode characters.
885
886The operating system, device drivers, C libraries, and Perl run-time
887system all conspire to let the programmer treat a single
888character (C<\n>) as the line terminator, irrespective of external
889representation.  On many operating systems, the native text file
890representation matches the internal representation, but on some
891platforms the external representation of C<\n> is made up of more than
892one character.
893
894All variants of Unix, Mac OS (old and new), and Stream_LF files on VMS use
895a single character to end each line in the external representation of text
896(even though that single character is CARRIAGE RETURN on old, pre-Darwin
897flavors of Mac OS, and is LINE FEED on Unix and most VMS files).  In other
898systems like OS/2, DOS, and the various flavors of MS-Windows, your program
899sees a C<\n> as a simple C<\cJ>, but what's stored in text files are the
900two characters C<\cM\cJ>.  That means that if you don't use
901L<C<binmode>|/binmode FILEHANDLE, LAYER> on these systems, C<\cM\cJ>
902sequences on disk will be converted to C<\n> on input, and any C<\n> in
903your program will be converted back to C<\cM\cJ> on output.  This is
904what you want for text files, but it can be disastrous for binary files.
905
906Another consequence of using L<C<binmode>|/binmode FILEHANDLE, LAYER>
907(on some systems) is that special end-of-file markers will be seen as
908part of the data stream.  For systems from the Microsoft family this
909means that, if your binary data contain C<\cZ>, the I/O subsystem will
910regard it as the end of the file, unless you use
911L<C<binmode>|/binmode FILEHANDLE, LAYER>.
912
913L<C<binmode>|/binmode FILEHANDLE, LAYER> is important not only for
914L<C<readline>|/readline EXPR> and L<C<print>|/print FILEHANDLE LIST>
915operations, but also when using
916L<C<read>|/read FILEHANDLE,SCALAR,LENGTH,OFFSET>,
917L<C<seek>|/seek FILEHANDLE,POSITION,WHENCE>,
918L<C<sysread>|/sysread FILEHANDLE,SCALAR,LENGTH,OFFSET>,
919L<C<syswrite>|/syswrite FILEHANDLE,SCALAR,LENGTH,OFFSET> and
920L<C<tell>|/tell FILEHANDLE> (see L<perlport> for more details).  See the
921L<C<$E<sol>>|perlvar/$E<sol>> and L<C<$\>|perlvar/$\> variables in
922L<perlvar> for how to manually set your input and output
923line-termination sequences.
924
925Portability issues: L<perlport/binmode>.
926
927=item bless REF,CLASSNAME
928X<bless>
929
930=item bless REF
931
932=for Pod::Functions create an object
933
934This function tells the thingy referenced by REF that it is now an object
935in the CLASSNAME package.  If CLASSNAME is an empty string, it is
936interpreted as referring to the C<main> package.
937If CLASSNAME is omitted, the current package
938is used.  Because a L<C<bless>|/bless REF,CLASSNAME> is often the last
939thing in a constructor, it returns the reference for convenience.
940Always use the two-argument version if a derived class might inherit the
941method doing the blessing.  See L<perlobj> for more about the blessing
942(and blessings) of objects.
943
944Consider always blessing objects in CLASSNAMEs that are mixed case.
945Namespaces with all lowercase names are considered reserved for
946Perl pragmas.  Builtin types have all uppercase names.  To prevent
947confusion, you may wish to avoid such package names as well.
948It is advised to avoid the class name C<0>, because much code erroneously
949uses the result of L<C<ref>|/ref EXPR> as a truth value.
950
951See L<perlmod/"Perl Modules">.
952
953=item break
954
955=for Pod::Functions +switch break out of a C<given> block
956
957Break out of a C<given> block.
958
959L<C<break>|/break> is available only if the
960L<C<"switch"> feature|feature/The 'switch' feature> is enabled or if it
961is prefixed with C<CORE::>. The
962L<C<"switch"> feature|feature/The 'switch' feature> is enabled
963automatically with a C<use v5.10> (or higher) declaration in the current
964scope.
965
966=item caller EXPR
967X<caller> X<call stack> X<stack> X<stack trace>
968
969=item caller
970
971=for Pod::Functions get context of the current subroutine call
972
973Returns the context of the current pure perl subroutine call.  In scalar
974context, returns the caller's package name if there I<is> a caller (that is, if
975we're in a subroutine or L<C<eval>|/eval EXPR> or
976L<C<require>|/require VERSION>) and the undefined value otherwise.
977caller never returns XS subs and they are skipped.  The next pure perl
978sub will appear instead of the XS sub in caller's return values.  In
979list context, caller returns
980
981       # 0         1          2
982    my ($package, $filename, $line) = caller;
983
984Like L<C<__FILE__>|/__FILE__> and L<C<__LINE__>|/__LINE__>, the filename and
985line number returned here may be altered by the mechanism described at
986L<perlsyn/"Plain Old Comments (Not!)">.
987
988With EXPR, it returns some extra information that the debugger uses to
989print a stack trace.  The value of EXPR indicates how many call frames
990to go back before the current one.
991
992    #  0         1          2      3            4
993 my ($package, $filename, $line, $subroutine, $hasargs,
994
995    #  5          6          7            8       9         10
996    $wantarray, $evaltext, $is_require, $hints, $bitmask, $hinthash)
997  = caller($i);
998
999Here, $subroutine is the function that the caller called (rather than the
1000function containing the caller).  Note that $subroutine may be C<(eval)> if
1001the frame is not a subroutine call, but an L<C<eval>|/eval EXPR>.  In
1002such a case additional elements $evaltext and C<$is_require> are set:
1003C<$is_require> is true if the frame is created by a
1004L<C<require>|/require VERSION> or L<C<use>|/use Module VERSION LIST>
1005statement, $evaltext contains the text of the C<eval EXPR> statement.
1006In particular, for an C<eval BLOCK> statement, $subroutine is C<(eval)>,
1007but $evaltext is undefined.  (Note also that each
1008L<C<use>|/use Module VERSION LIST> statement creates a
1009L<C<require>|/require VERSION> frame inside an C<eval EXPR> frame.)
1010$subroutine may also be C<(unknown)> if this particular subroutine
1011happens to have been deleted from the symbol table.  C<$hasargs> is true
1012if a new instance of L<C<@_>|perlvar/@_> was set up for the frame.
1013C<$hints> and C<$bitmask> contain pragmatic hints that the caller was
1014compiled with.  C<$hints> corresponds to L<C<$^H>|perlvar/$^H>, and
1015C<$bitmask> corresponds to
1016L<C<${^WARNING_BITS}>|perlvar/${^WARNING_BITS}>.  The C<$hints> and
1017C<$bitmask> values are subject to change between versions of Perl, and
1018are not meant for external use.
1019
1020C<$hinthash> is a reference to a hash containing the value of
1021L<C<%^H>|perlvar/%^H> when the caller was compiled, or
1022L<C<undef>|/undef EXPR> if L<C<%^H>|perlvar/%^H> was empty.  Do not
1023modify the values of this hash, as they are the actual values stored in
1024the optree.
1025
1026Note that the only types of call frames that are visible are subroutine
1027calls and C<eval>. Other forms of context, such as C<while> or C<foreach>
1028loops or C<try> blocks are not considered interesting to C<caller>, as they
1029do not alter the behaviour of the C<return> expression.
1030
1031Furthermore, when called from within the DB package in
1032list context, and with an argument, caller returns more
1033detailed information: it sets the list variable C<@DB::args> to be the
1034arguments with which the subroutine was invoked.
1035
1036Be aware that the optimizer might have optimized call frames away before
1037L<C<caller>|/caller EXPR> had a chance to get the information.  That
1038means that C<caller(N)> might not return information about the call
1039frame you expect it to, for C<< N > 1 >>.  In particular, C<@DB::args>
1040might have information from the previous time L<C<caller>|/caller EXPR>
1041was called.
1042
1043Be aware that setting C<@DB::args> is I<best effort>, intended for
1044debugging or generating backtraces, and should not be relied upon.  In
1045particular, as L<C<@_>|perlvar/@_> contains aliases to the caller's
1046arguments, Perl does not take a copy of L<C<@_>|perlvar/@_>, so
1047C<@DB::args> will contain modifications the subroutine makes to
1048L<C<@_>|perlvar/@_> or its contents, not the original values at call
1049time.  C<@DB::args>, like L<C<@_>|perlvar/@_>, does not hold explicit
1050references to its elements, so under certain cases its elements may have
1051become freed and reallocated for other variables or temporary values.
1052Finally, a side effect of the current implementation is that the effects
1053of C<shift @_> can I<normally> be undone (but not C<pop @_> or other
1054splicing, I<and> not if a reference to L<C<@_>|perlvar/@_> has been
1055taken, I<and> subject to the caveat about reallocated elements), so
1056C<@DB::args> is actually a hybrid of the current state and initial state
1057of L<C<@_>|perlvar/@_>.  Buyer beware.
1058
1059=item chdir EXPR
1060X<chdir>
1061X<cd>
1062X<directory, change>
1063
1064=item chdir FILEHANDLE
1065
1066=item chdir DIRHANDLE
1067
1068=item chdir
1069
1070=for Pod::Functions change your current working directory
1071
1072Changes the working directory to EXPR, if possible.  If EXPR is omitted,
1073changes to the directory specified by C<$ENV{HOME}>, if set; if not,
1074changes to the directory specified by C<$ENV{LOGDIR}>.  (Under VMS, the
1075variable C<$ENV{'SYS$LOGIN'}> is also checked, and used if it is set.)  If
1076neither is set, L<C<chdir>|/chdir EXPR> does nothing and fails.  It
1077returns true on success, false otherwise.  See the example under
1078L<C<die>|/die LIST>.
1079
1080On systems that support L<fchdir(2)>, you may pass a filehandle or
1081directory handle as the argument.  On systems that don't support L<fchdir(2)>,
1082passing handles raises an exception.
1083
1084=item chmod LIST
1085X<chmod> X<permission> X<mode>
1086
1087=for Pod::Functions changes the permissions on a list of files
1088
1089Changes the permissions of a list of files.  The first element of the
1090list must be the numeric mode, which should probably be an octal
1091number, and which definitely should I<not> be a string of octal digits:
1092C<0644> is okay, but C<"0644"> is not.  Returns the number of files
1093successfully changed.  See also L<C<oct>|/oct EXPR> if all you have is a
1094string.
1095
1096    my $cnt = chmod 0755, "foo", "bar";
1097    chmod 0755, @executables;
1098    my $mode = "0644"; chmod $mode, "foo";      # !!! sets mode to
1099                                                # --w----r-T
1100    my $mode = "0644"; chmod oct($mode), "foo"; # this is better
1101    my $mode = 0644;   chmod $mode, "foo";      # this is best
1102
1103On systems that support L<fchmod(2)>, you may pass filehandles among the
1104files.  On systems that don't support L<fchmod(2)>, passing filehandles raises
1105an exception.  Filehandles must be passed as globs or glob references to be
1106recognized; barewords are considered filenames.
1107
1108    open(my $fh, "<", "foo");
1109    my $perm = (stat $fh)[2] & 07777;
1110    chmod($perm | 0600, $fh);
1111
1112You can also import the symbolic C<S_I*> constants from the
1113L<C<Fcntl>|Fcntl> module:
1114
1115    use Fcntl qw( :mode );
1116    chmod S_IRWXU|S_IRGRP|S_IXGRP|S_IROTH|S_IXOTH, @executables;
1117    # Identical to the chmod 0755 of the example above.
1118
1119Portability issues: L<perlport/chmod>.
1120
1121=item chomp VARIABLE
1122X<chomp> X<INPUT_RECORD_SEPARATOR> X<$/> X<newline> X<eol>
1123
1124=item chomp( LIST )
1125
1126=item chomp
1127
1128=for Pod::Functions remove a trailing record separator from a string
1129
1130This safer version of L<C<chop>|/chop VARIABLE> removes any trailing
1131string that corresponds to the current value of
1132L<C<$E<sol>>|perlvar/$E<sol>> (also known as C<$INPUT_RECORD_SEPARATOR>
1133in the L<C<English>|English> module).  It returns the total
1134number of characters removed from all its arguments.  It's often used to
1135remove the newline from the end of an input record when you're worried
1136that the final record may be missing its newline.  When in paragraph
1137mode (C<$/ = ''>), it removes all trailing newlines from the string.
1138When in slurp mode (C<$/ = undef>) or fixed-length record mode
1139(L<C<$E<sol>>|perlvar/$E<sol>> is a reference to an integer or the like;
1140see L<perlvar>), L<C<chomp>|/chomp VARIABLE> won't remove anything.
1141If VARIABLE is omitted, it chomps L<C<$_>|perlvar/$_>.  Example:
1142
1143    while (<>) {
1144        chomp;  # avoid \n on last field
1145        my @array = split(/:/);
1146        # ...
1147    }
1148
1149If VARIABLE is a hash, it chomps the hash's values, but not its keys,
1150resetting the L<C<each>|/each HASH> iterator in the process.
1151
1152You can actually chomp anything that's an lvalue, including an assignment:
1153
1154    chomp(my $cwd = `pwd`);
1155    chomp(my $answer = <STDIN>);
1156
1157If you chomp a list, each element is chomped, and the total number of
1158characters removed is returned.
1159
1160Note that parentheses are necessary when you're chomping anything
1161that is not a simple variable.  This is because C<chomp $cwd = `pwd`;>
1162is interpreted as C<(chomp $cwd) = `pwd`;>, rather than as
1163C<chomp( $cwd = `pwd` )> which you might expect.  Similarly,
1164C<chomp $a, $b> is interpreted as C<chomp($a), $b> rather than
1165as C<chomp($a, $b)>.
1166
1167=item chop VARIABLE
1168X<chop>
1169
1170=item chop( LIST )
1171
1172=item chop
1173
1174=for Pod::Functions remove the last character from a string
1175
1176Chops off the last character of a string and returns the character
1177chopped.  It is much more efficient than C<s/.$//s> because it neither
1178scans nor copies the string.  If VARIABLE is omitted, chops
1179L<C<$_>|perlvar/$_>.
1180If VARIABLE is a hash, it chops the hash's values, but not its keys,
1181resetting the L<C<each>|/each HASH> iterator in the process.
1182
1183You can actually chop anything that's an lvalue, including an assignment.
1184
1185If you chop a list, each element is chopped.  Only the value of the
1186last L<C<chop>|/chop VARIABLE> is returned.
1187
1188Note that L<C<chop>|/chop VARIABLE> returns the last character.  To
1189return all but the last character, use C<substr($string, 0, -1)>.
1190
1191See also L<C<chomp>|/chomp VARIABLE>.
1192
1193=item chown LIST
1194X<chown> X<owner> X<user> X<group>
1195
1196=for Pod::Functions change the ownership on a list of files
1197
1198Changes the owner (and group) of a list of files.  The first two
1199elements of the list must be the I<numeric> uid and gid, in that
1200order.  A value of -1 in either position is interpreted by most
1201systems to leave that value unchanged.  Returns the number of files
1202successfully changed.
1203
1204    my $cnt = chown $uid, $gid, 'foo', 'bar';
1205    chown $uid, $gid, @filenames;
1206
1207On systems that support L<fchown(2)>, you may pass filehandles among the
1208files.  On systems that don't support L<fchown(2)>, passing filehandles raises
1209an exception.  Filehandles must be passed as globs or glob references to be
1210recognized; barewords are considered filenames.
1211
1212Here's an example that looks up nonnumeric uids in the passwd file:
1213
1214    print "User: ";
1215    chomp(my $user = <STDIN>);
1216    print "Files: ";
1217    chomp(my $pattern = <STDIN>);
1218
1219    my ($login,$pass,$uid,$gid) = getpwnam($user)
1220        or die "$user not in passwd file";
1221
1222    my @ary = glob($pattern);  # expand filenames
1223    chown $uid, $gid, @ary;
1224
1225On most systems, you are not allowed to change the ownership of the
1226file unless you're the superuser, although you should be able to change
1227the group to any of your secondary groups.  On insecure systems, these
1228restrictions may be relaxed, but this is not a portable assumption.
1229On POSIX systems, you can detect this condition this way:
1230
1231    use POSIX qw(sysconf _PC_CHOWN_RESTRICTED);
1232    my $can_chown_giveaway = ! sysconf(_PC_CHOWN_RESTRICTED);
1233
1234Portability issues: L<perlport/chown>.
1235
1236=item chr NUMBER
1237X<chr> X<character> X<ASCII> X<Unicode>
1238
1239=item chr
1240
1241=for Pod::Functions get character this number represents
1242
1243Returns the character represented by that NUMBER in the character set.
1244For example, C<chr(65)> is C<"A"> in either ASCII or Unicode, and
1245chr(0x263a) is a Unicode smiley face.
1246
1247Negative values give the Unicode replacement character (chr(0xfffd)),
1248except under the L<bytes> pragma, where the low eight bits of the value
1249(truncated to an integer) are used.
1250
1251If NUMBER is omitted, uses L<C<$_>|perlvar/$_>.
1252
1253For the reverse, use L<C<ord>|/ord EXPR>.
1254
1255Note that characters from 128 to 255 (inclusive) are by default
1256internally not encoded as UTF-8 for backward compatibility reasons.
1257
1258See L<perlunicode> for more about Unicode.
1259
1260=item chroot FILENAME
1261X<chroot> X<root>
1262
1263=item chroot
1264
1265=for Pod::Functions make directory new root for path lookups
1266
1267This function works like the system call by the same name: it makes the
1268named directory the new root directory for all further pathnames that
1269begin with a C</> by your process and all its children.  (It doesn't
1270change your current working directory, which is unaffected.)  For security
1271reasons, this call is restricted to the superuser.  If FILENAME is
1272omitted, does a L<C<chroot>|/chroot FILENAME> to L<C<$_>|perlvar/$_>.
1273
1274B<NOTE:>  It is mandatory for security to C<chdir("/")>
1275(L<C<chdir>|/chdir EXPR> to the root directory) immediately after a
1276L<C<chroot>|/chroot FILENAME>, otherwise the current working directory
1277may be outside of the new root.
1278
1279Portability issues: L<perlport/chroot>.
1280
1281=item close FILEHANDLE
1282X<close>
1283
1284=item close
1285
1286=for Pod::Functions close file (or pipe or socket) handle
1287
1288Closes the file or pipe associated with the filehandle, flushes the IO
1289buffers, and closes the system file descriptor.  Returns true if those
1290operations succeed and if no error was reported by any PerlIO
1291layer.  Closes the currently selected filehandle if the argument is
1292omitted.
1293
1294You don't have to close FILEHANDLE if you are immediately going to do
1295another L<C<open>|/open FILEHANDLE,MODE,EXPR> on it, because
1296L<C<open>|/open FILEHANDLE,MODE,EXPR> closes it for you.  (See
1297L<C<open>|/open FILEHANDLE,MODE,EXPR>.) However, an explicit
1298L<C<close>|/close FILEHANDLE> on an input file resets the line counter
1299(L<C<$.>|perlvar/$.>), while the implicit close done by
1300L<C<open>|/open FILEHANDLE,MODE,EXPR> does not.
1301
1302If the filehandle came from a piped open, L<C<close>|/close FILEHANDLE>
1303returns false if one of the other syscalls involved fails or if its
1304program exits with non-zero status.  If the only problem was that the
1305program exited non-zero, L<C<$!>|perlvar/$!> will be set to C<0>.
1306Closing a pipe also waits for the process executing on the pipe to
1307exit--in case you wish to look at the output of the pipe afterwards--and
1308implicitly puts the exit status value of that command into
1309L<C<$?>|perlvar/$?> and
1310L<C<${^CHILD_ERROR_NATIVE}>|perlvar/${^CHILD_ERROR_NATIVE}>.
1311
1312If there are multiple threads running, L<C<close>|/close FILEHANDLE> on
1313a filehandle from a piped open returns true without waiting for the
1314child process to terminate, if the filehandle is still open in another
1315thread.
1316
1317Closing the read end of a pipe before the process writing to it at the
1318other end is done writing results in the writer receiving a SIGPIPE.  If
1319the other end can't handle that, be sure to read all the data before
1320closing the pipe.
1321
1322Example:
1323
1324    open(OUTPUT, '|sort >foo')  # pipe to sort
1325        or die "Can't start sort: $!";
1326    #...                        # print stuff to output
1327    close OUTPUT                # wait for sort to finish
1328        or warn $! ? "Error closing sort pipe: $!"
1329                   : "Exit status $? from sort";
1330    open(INPUT, 'foo')          # get sort's results
1331        or die "Can't open 'foo' for input: $!";
1332
1333FILEHANDLE may be an expression whose value can be used as an indirect
1334filehandle, usually the real filehandle name or an autovivified handle.
1335
1336=item closedir DIRHANDLE
1337X<closedir>
1338
1339=for Pod::Functions close directory handle
1340
1341Closes a directory opened by L<C<opendir>|/opendir DIRHANDLE,EXPR> and
1342returns the success of that system call.
1343
1344=item connect SOCKET,NAME
1345X<connect>
1346
1347=for Pod::Functions connect to a remote socket
1348
1349Attempts to connect to a remote socket, just like L<connect(2)>.
1350Returns true if it succeeded, false otherwise.  NAME should be a
1351packed address of the appropriate type for the socket.  See the examples in
1352L<perlipc/"Sockets: Client/Server Communication">.
1353
1354=item continue BLOCK
1355X<continue>
1356
1357=item continue
1358
1359=for Pod::Functions optional trailing block in a while or foreach
1360
1361When followed by a BLOCK, L<C<continue>|/continue BLOCK> is actually a
1362flow control statement rather than a function.  If there is a
1363L<C<continue>|/continue BLOCK> BLOCK attached to a BLOCK (typically in a
1364C<while> or C<foreach>), it is always executed just before the
1365conditional is about to be evaluated again, just like the third part of
1366a C<for> loop in C.  Thus it can be used to increment a loop variable,
1367even when the loop has been continued via the L<C<next>|/next LABEL>
1368statement (which is similar to the C L<C<continue>|/continue BLOCK>
1369statement).
1370
1371L<C<last>|/last LABEL>, L<C<next>|/next LABEL>, or
1372L<C<redo>|/redo LABEL> may appear within a
1373L<C<continue>|/continue BLOCK> block; L<C<last>|/last LABEL> and
1374L<C<redo>|/redo LABEL> behave as if they had been executed within the
1375main block.  So will L<C<next>|/next LABEL>, but since it will execute a
1376L<C<continue>|/continue BLOCK> block, it may be more entertaining.
1377
1378    while (EXPR) {
1379        ### redo always comes here
1380        do_something;
1381    } continue {
1382        ### next always comes here
1383        do_something_else;
1384        # then back the top to re-check EXPR
1385    }
1386    ### last always comes here
1387
1388Omitting the L<C<continue>|/continue BLOCK> section is equivalent to
1389using an empty one, logically enough, so L<C<next>|/next LABEL> goes
1390directly back to check the condition at the top of the loop.
1391
1392When there is no BLOCK, L<C<continue>|/continue BLOCK> is a function
1393that falls through the current C<when> or C<default> block instead of
1394iterating a dynamically enclosing C<foreach> or exiting a lexically
1395enclosing C<given>.  In Perl 5.14 and earlier, this form of
1396L<C<continue>|/continue BLOCK> was only available when the
1397L<C<"switch"> feature|feature/The 'switch' feature> was enabled.  See
1398L<feature> and L<perlsyn/"Switch Statements"> for more information.
1399
1400=item cos EXPR
1401X<cos> X<cosine> X<acos> X<arccosine>
1402
1403=item cos
1404
1405=for Pod::Functions cosine function
1406
1407Returns the cosine of EXPR (expressed in radians).  If EXPR is omitted,
1408takes the cosine of L<C<$_>|perlvar/$_>.
1409
1410For the inverse cosine operation, you may use the
1411L<C<Math::Trig::acos>|Math::Trig> function, or use this relation:
1412
1413    sub acos { atan2( sqrt(1 - $_[0] * $_[0]), $_[0] ) }
1414
1415=item crypt PLAINTEXT,SALT
1416X<crypt> X<digest> X<hash> X<salt> X<plaintext> X<password>
1417X<decrypt> X<cryptography> X<passwd> X<encrypt>
1418
1419=for Pod::Functions one-way passwd-style encryption
1420
1421Creates a digest string exactly like the L<crypt(3)> function in the C
1422library (assuming that you actually have a version there that has not
1423been extirpated as a potential munition).
1424
1425L<C<crypt>|/crypt PLAINTEXT,SALT> is a one-way hash function.  The
1426PLAINTEXT and SALT are turned
1427into a short string, called a digest, which is returned.  The same
1428PLAINTEXT and SALT will always return the same string, but there is no
1429(known) way to get the original PLAINTEXT from the hash.  Small
1430changes in the PLAINTEXT or SALT will result in large changes in the
1431digest.
1432
1433There is no decrypt function.  This function isn't all that useful for
1434cryptography (for that, look for F<Crypt> modules on your nearby CPAN
1435mirror) and the name "crypt" is a bit of a misnomer.  Instead it is
1436primarily used to check if two pieces of text are the same without
1437having to transmit or store the text itself.  An example is checking
1438if a correct password is given.  The digest of the password is stored,
1439not the password itself.  The user types in a password that is
1440L<C<crypt>|/crypt PLAINTEXT,SALT>'d with the same salt as the stored
1441digest.  If the two digests match, the password is correct.
1442
1443When verifying an existing digest string you should use the digest as
1444the salt (like C<crypt($plain, $digest) eq $digest>).  The SALT used
1445to create the digest is visible as part of the digest.  This ensures
1446L<C<crypt>|/crypt PLAINTEXT,SALT> will hash the new string with the same
1447salt as the digest.  This allows your code to work with the standard
1448L<C<crypt>|/crypt PLAINTEXT,SALT> and with more exotic implementations.
1449In other words, assume nothing about the returned string itself nor
1450about how many bytes of SALT may matter.
1451
1452Traditionally the result is a string of 13 bytes: two first bytes of
1453the salt, followed by 11 bytes from the set C<[./0-9A-Za-z]>, and only
1454the first eight bytes of PLAINTEXT mattered.  But alternative
1455hashing schemes (like MD5), higher level security schemes (like C2),
1456and implementations on non-Unix platforms may produce different
1457strings.
1458
1459When choosing a new salt create a random two character string whose
1460characters come from the set C<[./0-9A-Za-z]> (like C<join '', ('.',
1461'/', 0..9, 'A'..'Z', 'a'..'z')[rand 64, rand 64]>).  This set of
1462characters is just a recommendation; the characters allowed in
1463the salt depend solely on your system's crypt library, and Perl can't
1464restrict what salts L<C<crypt>|/crypt PLAINTEXT,SALT> accepts.
1465
1466Here's an example that makes sure that whoever runs this program knows
1467their password:
1468
1469    my $pwd = (getpwuid($<))[1];
1470
1471    system "stty -echo";
1472    print "Password: ";
1473    chomp(my $word = <STDIN>);
1474    print "\n";
1475    system "stty echo";
1476
1477    if (crypt($word, $pwd) ne $pwd) {
1478        die "Sorry...\n";
1479    } else {
1480        print "ok\n";
1481    }
1482
1483Of course, typing in your own password to whoever asks you
1484for it is unwise.
1485
1486The L<C<crypt>|/crypt PLAINTEXT,SALT> function is unsuitable for hashing
1487large quantities of data, not least of all because you can't get the
1488information back.  Look at the L<Digest> module for more robust
1489algorithms.
1490
1491If using L<C<crypt>|/crypt PLAINTEXT,SALT> on a Unicode string (which
1492I<potentially> has characters with codepoints above 255), Perl tries to
1493make sense of the situation by trying to downgrade (a copy of) the
1494string back to an eight-bit byte string before calling
1495L<C<crypt>|/crypt PLAINTEXT,SALT> (on that copy).  If that works, good.
1496If not, L<C<crypt>|/crypt PLAINTEXT,SALT> dies with
1497L<C<Wide character in crypt>|perldiag/Wide character in %s>.
1498
1499Portability issues: L<perlport/crypt>.
1500
1501=item dbmclose HASH
1502X<dbmclose>
1503
1504=for Pod::Functions breaks binding on a tied dbm file
1505
1506[This function has been largely superseded by the
1507L<C<untie>|/untie VARIABLE> function.]
1508
1509Breaks the binding between a DBM file and a hash.
1510
1511Portability issues: L<perlport/dbmclose>.
1512
1513=item dbmopen HASH,DBNAME,MASK
1514X<dbmopen> X<dbm> X<ndbm> X<sdbm> X<gdbm>
1515
1516=for Pod::Functions create binding on a tied dbm file
1517
1518[This function has been largely superseded by the
1519L<C<tie>|/tie VARIABLE,CLASSNAME,LIST> function.]
1520
1521This binds a L<dbm(3)>, L<ndbm(3)>, L<sdbm(3)>, L<gdbm(3)>, or Berkeley
1522DB file to a hash.  HASH is the name of the hash.  (Unlike normal
1523L<C<open>|/open FILEHANDLE,MODE,EXPR>, the first argument is I<not> a
1524filehandle, even though it looks like one).  DBNAME is the name of the
1525database (without the F<.dir> or F<.pag> extension if any).  If the
1526database does not exist, it is created with protection specified by MASK
1527(as modified by the L<C<umask>|/umask EXPR>).  To prevent creation of
1528the database if it doesn't exist, you may specify a MODE of 0, and the
1529function will return a false value if it can't find an existing
1530database.  If your system supports only the older DBM functions, you may
1531make only one L<C<dbmopen>|/dbmopen HASH,DBNAME,MASK> call in your
1532program.  In older versions of Perl, if your system had neither DBM nor
1533ndbm, calling L<C<dbmopen>|/dbmopen HASH,DBNAME,MASK> produced a fatal
1534error; it now falls back to L<sdbm(3)>.
1535
1536If you don't have write access to the DBM file, you can only read hash
1537variables, not set them.  If you want to test whether you can write,
1538either use file tests or try setting a dummy hash entry inside an
1539L<C<eval>|/eval EXPR> to trap the error.
1540
1541Note that functions such as L<C<keys>|/keys HASH> and
1542L<C<values>|/values HASH> may return huge lists when used on large DBM
1543files.  You may prefer to use the L<C<each>|/each HASH> function to
1544iterate over large DBM files.  Example:
1545
1546    # print out history file offsets
1547    dbmopen(%HIST,'/usr/lib/news/history',0666);
1548    while (($key,$val) = each %HIST) {
1549        print $key, ' = ', unpack('L',$val), "\n";
1550    }
1551    dbmclose(%HIST);
1552
1553See also L<AnyDBM_File> for a more general description of the pros and
1554cons of the various dbm approaches, as well as L<DB_File> for a particularly
1555rich implementation.
1556
1557You can control which DBM library you use by loading that library
1558before you call L<C<dbmopen>|/dbmopen HASH,DBNAME,MASK>:
1559
1560    use DB_File;
1561    dbmopen(%NS_Hist, "$ENV{HOME}/.netscape/history.db")
1562        or die "Can't open netscape history file: $!";
1563
1564Portability issues: L<perlport/dbmopen>.
1565
1566=item defined EXPR
1567X<defined> X<undef> X<undefined>
1568
1569=item defined
1570
1571=for Pod::Functions test whether a value, variable, or function is defined
1572
1573Returns a Boolean value telling whether EXPR has a value other than the
1574undefined value L<C<undef>|/undef EXPR>.  If EXPR is not present,
1575L<C<$_>|perlvar/$_> is checked.
1576
1577Many operations return L<C<undef>|/undef EXPR> to indicate failure, end
1578of file, system error, uninitialized variable, and other exceptional
1579conditions.  This function allows you to distinguish
1580L<C<undef>|/undef EXPR> from other values.  (A simple Boolean test will
1581not distinguish among L<C<undef>|/undef EXPR>, zero, the empty string,
1582and C<"0">, which are all equally false.)  Note that since
1583L<C<undef>|/undef EXPR> is a valid scalar, its presence doesn't
1584I<necessarily> indicate an exceptional condition: L<C<pop>|/pop ARRAY>
1585returns L<C<undef>|/undef EXPR> when its argument is an empty array,
1586I<or> when the element to return happens to be L<C<undef>|/undef EXPR>.
1587
1588You may also use C<defined(&func)> to check whether subroutine C<func>
1589has ever been defined.  The return value is unaffected by any forward
1590declarations of C<func>.  A subroutine that is not defined
1591may still be callable: its package may have an C<AUTOLOAD> method that
1592makes it spring into existence the first time that it is called; see
1593L<perlsub>.
1594
1595Use of L<C<defined>|/defined EXPR> on aggregates (hashes and arrays) is
1596no longer supported. It used to report whether memory for that
1597aggregate had ever been allocated.  You should instead use a simple
1598test for size:
1599
1600    if (@an_array) { print "has array elements\n" }
1601    if (%a_hash)   { print "has hash members\n"   }
1602
1603When used on a hash element, it tells you whether the value is defined,
1604not whether the key exists in the hash.  Use L<C<exists>|/exists EXPR>
1605for the latter purpose.
1606
1607Examples:
1608
1609    print if defined $switch{D};
1610    print "$val\n" while defined($val = pop(@ary));
1611    die "Can't readlink $sym: $!"
1612        unless defined($value = readlink $sym);
1613    sub foo { defined &$bar ? $bar->(@_) : die "No bar"; }
1614    $debugging = 0 unless defined $debugging;
1615
1616Note:  Many folks tend to overuse L<C<defined>|/defined EXPR> and are
1617then surprised to discover that the number C<0> and C<""> (the
1618zero-length string) are, in fact, defined values.  For example, if you
1619say
1620
1621    "ab" =~ /a(.*)b/;
1622
1623The pattern match succeeds and C<$1> is defined, although it
1624matched "nothing".  It didn't really fail to match anything.  Rather, it
1625matched something that happened to be zero characters long.  This is all
1626very above-board and honest.  When a function returns an undefined value,
1627it's an admission that it couldn't give you an honest answer.  So you
1628should use L<C<defined>|/defined EXPR> only when questioning the
1629integrity of what you're trying to do.  At other times, a simple
1630comparison to C<0> or C<""> is what you want.
1631
1632See also L<C<undef>|/undef EXPR>, L<C<exists>|/exists EXPR>,
1633L<C<ref>|/ref EXPR>.
1634
1635=item delete EXPR
1636X<delete>
1637
1638=for Pod::Functions deletes a value from a hash
1639
1640Given an expression that specifies an element or slice of a hash,
1641L<C<delete>|/delete EXPR> deletes the specified elements from that hash
1642so that L<C<exists>|/exists EXPR> on that element no longer returns
1643true.  Setting a hash element to the undefined value does not remove its
1644key, but deleting it does; see L<C<exists>|/exists EXPR>.
1645
1646In list context, usually returns the value or values deleted, or the last such
1647element in scalar context.  The return list's length corresponds to that of
1648the argument list: deleting non-existent elements returns the undefined value
1649in their corresponding positions. Since Perl 5.28, a
1650L<keyE<sol>value hash slice|perldata/KeyE<sol>Value Hash Slices> can be passed
1651to C<delete>, and the return value is a list of key/value pairs (two elements
1652for each item deleted from the hash).
1653
1654L<C<delete>|/delete EXPR> may also be used on arrays and array slices,
1655but its behavior is less straightforward.  Although
1656L<C<exists>|/exists EXPR> will return false for deleted entries,
1657deleting array elements never changes indices of existing values; use
1658L<C<shift>|/shift ARRAY> or L<C<splice>|/splice
1659ARRAY,OFFSET,LENGTH,LIST> for that.  However, if any deleted elements
1660fall at the end of an array, the array's size shrinks to the position of
1661the highest element that still tests true for L<C<exists>|/exists EXPR>,
1662or to 0 if none do.  In other words, an array won't have trailing
1663nonexistent elements after a delete.
1664
1665B<WARNING:> Calling L<C<delete>|/delete EXPR> on array values is
1666strongly discouraged.  The
1667notion of deleting or checking the existence of Perl array elements is not
1668conceptually coherent, and can lead to surprising behavior.
1669
1670Deleting from L<C<%ENV>|perlvar/%ENV> modifies the environment.
1671Deleting from a hash tied to a DBM file deletes the entry from the DBM
1672file.  Deleting from a L<C<tied>|/tied VARIABLE> hash or array may not
1673necessarily return anything; it depends on the implementation of the
1674L<C<tied>|/tied VARIABLE> package's DELETE method, which may do whatever
1675it pleases.
1676
1677The C<delete local EXPR> construct localizes the deletion to the current
1678block at run time.  Until the block exits, elements locally deleted
1679temporarily no longer exist.  See L<perlsub/"Localized deletion of elements
1680of composite types">.
1681
1682    my %hash = (foo => 11, bar => 22, baz => 33);
1683    my $scalar = delete $hash{foo};         # $scalar is 11
1684    $scalar = delete @hash{qw(foo bar)}; # $scalar is 22
1685    my @array  = delete @hash{qw(foo baz)}; # @array  is (undef,33)
1686
1687The following (inefficiently) deletes all the values of %HASH and @ARRAY:
1688
1689    foreach my $key (keys %HASH) {
1690        delete $HASH{$key};
1691    }
1692
1693    foreach my $index (0 .. $#ARRAY) {
1694        delete $ARRAY[$index];
1695    }
1696
1697And so do these:
1698
1699    delete @HASH{keys %HASH};
1700
1701    delete @ARRAY[0 .. $#ARRAY];
1702
1703But both are slower than assigning the empty list
1704or undefining %HASH or @ARRAY, which is the customary
1705way to empty out an aggregate:
1706
1707    %HASH = ();     # completely empty %HASH
1708    undef %HASH;    # forget %HASH ever existed
1709
1710    @ARRAY = ();    # completely empty @ARRAY
1711    undef @ARRAY;   # forget @ARRAY ever existed
1712
1713The EXPR can be arbitrarily complicated provided its
1714final operation is an element or slice of an aggregate:
1715
1716    delete $ref->[$x][$y]{$key};
1717    delete $ref->[$x][$y]->@{$key1, $key2, @morekeys};
1718
1719    delete $ref->[$x][$y][$index];
1720    delete $ref->[$x][$y]->@[$index1, $index2, @moreindices];
1721
1722=item die LIST
1723X<die> X<throw> X<exception> X<raise> X<$@> X<abort>
1724
1725=for Pod::Functions raise an exception or bail out
1726
1727L<C<die>|/die LIST> raises an exception.  Inside an L<C<eval>|/eval EXPR>
1728the exception is stuffed into L<C<$@>|perlvar/$@> and the L<C<eval>|/eval
1729EXPR> is terminated with the undefined value.  If the exception is
1730outside of all enclosing L<C<eval>|/eval EXPR>s, then the uncaught
1731exception is printed to C<STDERR> and perl exits with an exit code
1732indicating failure.  If you need to exit the process with a specific
1733exit code, see L<C<exit>|/exit EXPR>.
1734
1735Equivalent examples:
1736
1737    die "Can't cd to spool: $!\n" unless chdir '/usr/spool/news';
1738    chdir '/usr/spool/news' or die "Can't cd to spool: $!\n"
1739
1740Most of the time, C<die> is called with a string to use as the exception.
1741You may either give a single non-reference operand to serve as the
1742exception, or a list of two or more items, which will be stringified
1743and concatenated to make the exception.
1744
1745If the string exception does not end in a newline, the current
1746script line number and input line number (if any) and a newline
1747are appended to it.  Note that the "input line number" (also
1748known as "chunk") is subject to whatever notion of "line" happens to
1749be currently in effect, and is also available as the special variable
1750L<C<$.>|perlvar/$.>.  See L<perlvar/"$/"> and L<perlvar/"$.">.
1751
1752Hint: sometimes appending C<", stopped"> to your message will cause it
1753to make better sense when the string C<"at foo line 123"> is appended.
1754Suppose you are running script "canasta".
1755
1756    die "/etc/games is no good";
1757    die "/etc/games is no good, stopped";
1758
1759produce, respectively
1760
1761    /etc/games is no good at canasta line 123.
1762    /etc/games is no good, stopped at canasta line 123.
1763
1764If LIST was empty or made an empty string, and L<C<$@>|perlvar/$@>
1765already contains an exception value (typically from a previous
1766L<C<eval>|/eval EXPR>), then that value is reused after
1767appending C<"\t...propagated">.  This is useful for propagating exceptions:
1768
1769    eval { ... };
1770    die unless $@ =~ /Expected exception/;
1771
1772If LIST was empty or made an empty string,
1773and L<C<$@>|perlvar/$@> contains an object
1774reference that has a C<PROPAGATE> method, that method will be called
1775with additional file and line number parameters.  The return value
1776replaces the value in L<C<$@>|perlvar/$@>;  i.e., as if
1777C<< $@ = eval { $@->PROPAGATE(__FILE__, __LINE__) }; >> were called.
1778
1779If LIST was empty or made an empty string, and L<C<$@>|perlvar/$@>
1780is also empty, then the string C<"Died"> is used.
1781
1782You can also call L<C<die>|/die LIST> with a reference argument, and if
1783this is trapped within an L<C<eval>|/eval EXPR>, L<C<$@>|perlvar/$@>
1784contains that reference.  This permits more elaborate exception handling
1785using objects that maintain arbitrary state about the exception.  Such a
1786scheme is sometimes preferable to matching particular string values of
1787L<C<$@>|perlvar/$@> with regular expressions.
1788
1789Because Perl stringifies uncaught exception messages before display,
1790you'll probably want to overload stringification operations on
1791exception objects.  See L<overload> for details about that.
1792The stringified message should be non-empty, and should end in a newline,
1793in order to fit in with the treatment of string exceptions.
1794Also, because an exception object reference cannot be stringified
1795without destroying it, Perl doesn't attempt to append location or other
1796information to a reference exception.  If you want location information
1797with a complex exception object, you'll have to arrange to put the
1798location information into the object yourself.
1799
1800Because L<C<$@>|perlvar/$@> is a global variable, be careful that
1801analyzing an exception caught by C<eval> doesn't replace the reference
1802in the global variable.  It's
1803easiest to make a local copy of the reference before any manipulations.
1804Here's an example:
1805
1806    use Scalar::Util "blessed";
1807
1808    eval { ... ; die Some::Module::Exception->new( FOO => "bar" ) };
1809    if (my $ev_err = $@) {
1810        if (blessed($ev_err)
1811            && $ev_err->isa("Some::Module::Exception")) {
1812            # handle Some::Module::Exception
1813        }
1814        else {
1815            # handle all other possible exceptions
1816        }
1817    }
1818
1819If an uncaught exception results in interpreter exit, the exit code is
1820determined from the values of L<C<$!>|perlvar/$!> and
1821L<C<$?>|perlvar/$?> with this pseudocode:
1822
1823    exit $! if $!;              # errno
1824    exit $? >> 8 if $? >> 8;    # child exit status
1825    exit 255;                   # last resort
1826
1827As with L<C<exit>|/exit EXPR>, L<C<$?>|perlvar/$?> is set prior to
1828unwinding the call stack; any C<DESTROY> or C<END> handlers can then
1829alter this value, and thus Perl's exit code.
1830
1831The intent is to squeeze as much possible information about the likely cause
1832into the limited space of the system exit code.  However, as
1833L<C<$!>|perlvar/$!> is the value of C's C<errno>, which can be set by
1834any system call, this means that the value of the exit code used by
1835L<C<die>|/die LIST> can be non-predictable, so should not be relied
1836upon, other than to be non-zero.
1837
1838You can arrange for a callback to be run just before the
1839L<C<die>|/die LIST> does its deed, by setting the
1840L<C<$SIG{__DIE__}>|perlvar/%SIG> hook.  The associated handler is called
1841with the exception as an argument, and can change the exception,
1842if it sees fit, by
1843calling L<C<die>|/die LIST> again.  See L<perlvar/%SIG> for details on
1844setting L<C<%SIG>|perlvar/%SIG> entries, and L<C<eval>|/eval EXPR> for some
1845examples.  Although this feature was to be run only right before your
1846program was to exit, this is not currently so: the
1847L<C<$SIG{__DIE__}>|perlvar/%SIG> hook is currently called even inside
1848L<C<eval>|/eval EXPR>ed blocks/strings!  If one wants the hook to do
1849nothing in such situations, put
1850
1851    die @_ if $^S;
1852
1853as the first line of the handler (see L<perlvar/$^S>).  Because
1854this promotes strange action at a distance, this counterintuitive
1855behavior may be fixed in a future release.
1856
1857See also L<C<exit>|/exit EXPR>, L<C<warn>|/warn LIST>, and the L<Carp>
1858module.
1859
1860=item do BLOCK
1861X<do> X<block>
1862
1863=for Pod::Functions turn a BLOCK into a TERM
1864
1865Not really a function.  Returns the value of the last command in the
1866sequence of commands indicated by BLOCK.  When modified by the C<while> or
1867C<until> loop modifier, executes the BLOCK once before testing the loop
1868condition.  (On other statements the loop modifiers test the conditional
1869first.)
1870
1871C<do BLOCK> does I<not> count as a loop, so the loop control statements
1872L<C<next>|/next LABEL>, L<C<last>|/last LABEL>, or
1873L<C<redo>|/redo LABEL> cannot be used to leave or restart the block.
1874See L<perlsyn> for alternative strategies.
1875
1876=item do EXPR
1877X<do>
1878
1879Uses the value of EXPR as a filename and executes the contents of the
1880file as a Perl script:
1881
1882    # load the exact specified file (./ and ../ special-cased)
1883    do '/foo/stat.pl';
1884    do './stat.pl';
1885    do '../foo/stat.pl';
1886
1887    # search for the named file within @INC
1888    do 'stat.pl';
1889    do 'foo/stat.pl';
1890
1891C<do './stat.pl'> is largely like
1892
1893    eval `cat stat.pl`;
1894
1895except that it's more concise, runs no external processes, and keeps
1896track of the current filename for error messages. It also differs in that
1897code evaluated with C<do FILE> cannot see lexicals in the enclosing
1898scope; C<eval STRING> does.  It's the same, however, in that it does
1899reparse the file every time you call it, so you probably don't want
1900to do this inside a loop.
1901
1902Using C<do> with a relative path (except for F<./> and F<../>), like
1903
1904    do 'foo/stat.pl';
1905
1906will search the L<C<@INC>|perlvar/@INC> directories, and update
1907L<C<%INC>|perlvar/%INC> if the file is found.  See L<perlvar/@INC>
1908and L<perlvar/%INC> for these variables. In particular, note that
1909whilst historically L<C<@INC>|perlvar/@INC> contained '.' (the
1910current directory) making these two cases equivalent, that is no
1911longer necessarily the case, as '.' is not included in C<@INC> by default
1912in perl versions 5.26.0 onwards. Instead, perl will now warn:
1913
1914    do "stat.pl" failed, '.' is no longer in @INC;
1915    did you mean do "./stat.pl"?
1916
1917If L<C<do>|/do EXPR> can read the file but cannot compile it, it
1918returns L<C<undef>|/undef EXPR> and sets an error message in
1919L<C<$@>|perlvar/$@>.  If L<C<do>|/do EXPR> cannot read the file, it
1920returns undef and sets L<C<$!>|perlvar/$!> to the error.  Always check
1921L<C<$@>|perlvar/$@> first, as compilation could fail in a way that also
1922sets L<C<$!>|perlvar/$!>.  If the file is successfully compiled,
1923L<C<do>|/do EXPR> returns the value of the last expression evaluated.
1924
1925Inclusion of library modules is better done with the
1926L<C<use>|/use Module VERSION LIST> and L<C<require>|/require VERSION>
1927operators, which also do automatic error checking and raise an exception
1928if there's a problem.
1929
1930You might like to use L<C<do>|/do EXPR> to read in a program
1931configuration file.  Manual error checking can be done this way:
1932
1933    # Read in config files: system first, then user.
1934    # Beware of using relative pathnames here.
1935    for $file ("/share/prog/defaults.rc",
1936               "$ENV{HOME}/.someprogrc")
1937    {
1938        unless ($return = do $file) {
1939            warn "couldn't parse $file: $@" if $@;
1940            warn "couldn't do $file: $!"    unless defined $return;
1941            warn "couldn't run $file"       unless $return;
1942        }
1943    }
1944
1945=item dump LABEL
1946X<dump> X<core> X<undump>
1947
1948=item dump EXPR
1949
1950=item dump
1951
1952=for Pod::Functions create an immediate core dump
1953
1954This function causes an immediate core dump.  See also the B<-u>
1955command-line switch in L<perlrun|perlrun/-u>, which does the same thing.
1956Primarily this is so that you can use the B<undump> program (not
1957supplied) to turn your core dump into an executable binary after
1958having initialized all your variables at the beginning of the
1959program.  When the new binary is executed it will begin by executing
1960a C<goto LABEL> (with all the restrictions that L<C<goto>|/goto LABEL>
1961suffers).
1962Think of it as a goto with an intervening core dump and reincarnation.
1963If C<LABEL> is omitted, restarts the program from the top.  The
1964C<dump EXPR> form, available starting in Perl 5.18.0, allows a name to be
1965computed at run time, being otherwise identical to C<dump LABEL>.
1966
1967B<WARNING>: Any files opened at the time of the dump will I<not>
1968be open any more when the program is reincarnated, with possible
1969resulting confusion by Perl.
1970
1971This function is now largely obsolete, mostly because it's very hard to
1972convert a core file into an executable.  As of Perl 5.30, it must be invoked
1973as C<CORE::dump()>.
1974
1975Unlike most named operators, this has the same precedence as assignment.
1976It is also exempt from the looks-like-a-function rule, so
1977C<dump ("foo")."bar"> will cause "bar" to be part of the argument to
1978L<C<dump>|/dump LABEL>.
1979
1980Portability issues: L<perlport/dump>.
1981
1982=item each HASH
1983X<each> X<hash, iterator>
1984
1985=item each ARRAY
1986X<array, iterator>
1987
1988=for Pod::Functions retrieve the next key/value pair from a hash
1989
1990When called on a hash in list context, returns a 2-element list
1991consisting of the key and value for the next element of a hash.  In Perl
19925.12 and later only, it will also return the index and value for the next
1993element of an array so that you can iterate over it; older Perls consider
1994this a syntax error.  When called in scalar context, returns only the key
1995(not the value) in a hash, or the index in an array.
1996
1997Hash entries are returned in an apparently random order.  The actual random
1998order is specific to a given hash; the exact same series of operations
1999on two hashes may result in a different order for each hash.  Any insertion
2000into the hash may change the order, as will any deletion, with the exception
2001that the most recent key returned by L<C<each>|/each HASH> or
2002L<C<keys>|/keys HASH> may be deleted without changing the order.  So
2003long as a given hash is unmodified you may rely on
2004L<C<keys>|/keys HASH>, L<C<values>|/values HASH> and
2005L<C<each>|/each HASH> to repeatedly return the same order
2006as each other.  See L<perlsec/"Algorithmic Complexity Attacks"> for
2007details on why hash order is randomized.  Aside from the guarantees
2008provided here the exact details of Perl's hash algorithm and the hash
2009traversal order are subject to change in any release of Perl.
2010
2011After L<C<each>|/each HASH> has returned all entries from the hash or
2012array, the next call to L<C<each>|/each HASH> returns the empty list in
2013list context and L<C<undef>|/undef EXPR> in scalar context; the next
2014call following I<that> one restarts iteration.  Each hash or array has
2015its own internal iterator, accessed by L<C<each>|/each HASH>,
2016L<C<keys>|/keys HASH>, and L<C<values>|/values HASH>.  The iterator is
2017implicitly reset when L<C<each>|/each HASH> has reached the end as just
2018described; it can be explicitly reset by calling L<C<keys>|/keys HASH>
2019or L<C<values>|/values HASH> on the hash or array, or by referencing
2020the hash (but not array) in list context.  If you add or delete
2021a hash's elements while iterating over it, the effect on the iterator is
2022unspecified; for example, entries may be skipped or duplicated--so don't
2023do that.  Exception: It is always safe to delete the item most recently
2024returned by L<C<each>|/each HASH>, so the following code works properly:
2025
2026    while (my ($key, $value) = each %hash) {
2027        print $key, "\n";
2028        delete $hash{$key};   # This is safe
2029    }
2030
2031Tied hashes may have a different ordering behaviour to perl's hash
2032implementation.
2033
2034The iterator used by C<each> is attached to the hash or array, and is
2035shared between all iteration operations applied to the same hash or array.
2036Thus all uses of C<each> on a single hash or array advance the same
2037iterator location.  All uses of C<each> are also subject to having the
2038iterator reset by any use of C<keys> or C<values> on the same hash or
2039array, or by the hash (but not array) being referenced in list context.
2040This makes C<each>-based loops quite fragile: it is easy to arrive at
2041such a loop with the iterator already part way through the object, or to
2042accidentally clobber the iterator state during execution of the loop body.
2043It's easy enough to explicitly reset the iterator before starting a loop,
2044but there is no way to insulate the iterator state used by a loop from
2045the iterator state used by anything else that might execute during the
2046loop body.  To avoid these problems, use a C<foreach> loop rather than
2047C<while>-C<each>.
2048
2049This extends to using C<each> on the result of an anonymous hash or
2050array constructor.  A new underlying array or hash is created each
2051time so each will always start iterating from scratch, eg:
2052
2053  # loops forever
2054  while (my ($key, $value) = each @{ +{ a => 1 } }) {
2055      print "$key=$value\n";
2056  }
2057
2058This prints out your environment like the L<printenv(1)> program,
2059but in a different order:
2060
2061    while (my ($key,$value) = each %ENV) {
2062        print "$key=$value\n";
2063    }
2064
2065Starting with Perl 5.14, an experimental feature allowed
2066L<C<each>|/each HASH> to take a scalar expression. This experiment has
2067been deemed unsuccessful, and was removed as of Perl 5.24.
2068
2069As of Perl 5.18 you can use a bare L<C<each>|/each HASH> in a C<while>
2070loop, which will set L<C<$_>|perlvar/$_> on every iteration.
2071If either an C<each> expression or an explicit assignment of an C<each>
2072expression to a scalar is used as a C<while>/C<for> condition, then
2073the condition actually tests for definedness of the expression's value,
2074not for its regular truth value.
2075
2076    while (each %ENV) {
2077	print "$_=$ENV{$_}\n";
2078    }
2079
2080To avoid confusing would-be users of your code who are running earlier
2081versions of Perl with mysterious syntax errors, put this sort of thing at
2082the top of your file to signal that your code will work I<only> on Perls of
2083a recent vintage:
2084
2085    use v5.12;	# so keys/values/each work on arrays
2086    use v5.18;	# so each assigns to $_ in a lone while test
2087
2088See also L<C<keys>|/keys HASH>, L<C<values>|/values HASH>, and
2089L<C<sort>|/sort SUBNAME LIST>.
2090
2091=item eof FILEHANDLE
2092X<eof>
2093X<end of file>
2094X<end-of-file>
2095
2096=item eof ()
2097
2098=item eof
2099
2100=for Pod::Functions test a filehandle for its end
2101
2102Returns 1 if the next read on FILEHANDLE will return end of file I<or> if
2103FILEHANDLE is not open.  FILEHANDLE may be an expression whose value
2104gives the real filehandle.  (Note that this function actually
2105reads a character and then C<ungetc>s it, so isn't useful in an
2106interactive context.)  Do not read from a terminal file (or call
2107C<eof(FILEHANDLE)> on it) after end-of-file is reached.  File types such
2108as terminals may lose the end-of-file condition if you do.
2109
2110An L<C<eof>|/eof FILEHANDLE> without an argument uses the last file
2111read.  Using L<C<eof()>|/eof FILEHANDLE> with empty parentheses is
2112different.  It refers to the pseudo file formed from the files listed on
2113the command line and accessed via the C<< <> >> operator.  Since
2114C<< <> >> isn't explicitly opened, as a normal filehandle is, an
2115L<C<eof()>|/eof FILEHANDLE> before C<< <> >> has been used will cause
2116L<C<@ARGV>|perlvar/@ARGV> to be examined to determine if input is
2117available.   Similarly, an L<C<eof()>|/eof FILEHANDLE> after C<< <> >>
2118has returned end-of-file will assume you are processing another
2119L<C<@ARGV>|perlvar/@ARGV> list, and if you haven't set
2120L<C<@ARGV>|perlvar/@ARGV>, will read input from C<STDIN>; see
2121L<perlop/"I/O Operators">.
2122
2123In a C<< while (<>) >> loop, L<C<eof>|/eof FILEHANDLE> or C<eof(ARGV)>
2124can be used to detect the end of each file, whereas
2125L<C<eof()>|/eof FILEHANDLE> will detect the end of the very last file
2126only.  Examples:
2127
2128    # reset line numbering on each input file
2129    while (<>) {
2130        next if /^\s*#/;  # skip comments
2131        print "$.\t$_";
2132    } continue {
2133        close ARGV if eof;  # Not eof()!
2134    }
2135
2136    # insert dashes just before last line of last file
2137    while (<>) {
2138        if (eof()) {  # check for end of last file
2139            print "--------------\n";
2140        }
2141        print;
2142        last if eof();     # needed if we're reading from a terminal
2143    }
2144
2145Practical hint: you almost never need to use L<C<eof>|/eof FILEHANDLE>
2146in Perl, because the input operators typically return L<C<undef>|/undef
2147EXPR> when they run out of data or encounter an error.
2148
2149=item eval EXPR
2150X<eval> X<try> X<catch> X<evaluate> X<parse> X<execute>
2151X<error, handling> X<exception, handling>
2152
2153=item eval BLOCK
2154
2155=item eval
2156
2157=for Pod::Functions catch exceptions or compile and run code
2158
2159C<eval> in all its forms is used to execute a little Perl program,
2160trapping any errors encountered so they don't crash the calling program.
2161
2162Plain C<eval> with no argument is just C<eval EXPR>, where the
2163expression is understood to be contained in L<C<$_>|perlvar/$_>.  Thus
2164there are only two real C<eval> forms; the one with an EXPR is often
2165called "string eval".  In a string eval, the value of the expression
2166(which is itself determined within scalar context) is first parsed, and
2167if there were no errors, executed as a block within the lexical context
2168of the current Perl program.  This form is typically used to delay
2169parsing and subsequent execution of the text of EXPR until run time.
2170Note that the value is parsed every time the C<eval> executes.
2171
2172The other form is called "block eval".  It is less general than string
2173eval, but the code within the BLOCK is parsed only once (at the same
2174time the code surrounding the C<eval> itself was parsed) and executed
2175within the context of the current Perl program.  This form is typically
2176used to trap exceptions more efficiently than the first, while also
2177providing the benefit of checking the code within BLOCK at compile time.
2178BLOCK is parsed and compiled just once.  Since errors are trapped, it
2179often is used to check if a given feature is available.
2180
2181In both forms, the value returned is the value of the last expression
2182evaluated inside the mini-program; a return statement may also be used, just
2183as with subroutines.  The expression providing the return value is evaluated
2184in void, scalar, or list context, depending on the context of the
2185C<eval> itself.  See L<C<wantarray>|/wantarray> for more
2186on how the evaluation context can be determined.
2187
2188If there is a syntax error or runtime error, or a L<C<die>|/die LIST>
2189statement is executed, C<eval> returns
2190L<C<undef>|/undef EXPR> in scalar context, or an empty list in list
2191context, and L<C<$@>|perlvar/$@> is set to the error message.  (Prior to
21925.16, a bug caused L<C<undef>|/undef EXPR> to be returned in list
2193context for syntax errors, but not for runtime errors.) If there was no
2194error, L<C<$@>|perlvar/$@> is set to the empty string.  A control flow
2195operator like L<C<last>|/last LABEL> or L<C<goto>|/goto LABEL> can
2196bypass the setting of L<C<$@>|perlvar/$@>.  Beware that using
2197C<eval> neither silences Perl from printing warnings to
2198STDERR, nor does it stuff the text of warning messages into
2199L<C<$@>|perlvar/$@>.  To do either of those, you have to use the
2200L<C<$SIG{__WARN__}>|perlvar/%SIG> facility, or turn off warnings inside
2201the BLOCK or EXPR using S<C<no warnings 'all'>>.  See
2202L<C<warn>|/warn LIST>, L<perlvar>, and L<warnings>.
2203
2204Note that, because C<eval> traps otherwise-fatal errors,
2205it is useful for determining whether a particular feature (such as
2206L<C<socket>|/socket SOCKET,DOMAIN,TYPE,PROTOCOL> or
2207L<C<symlink>|/symlink OLDFILE,NEWFILE>) is implemented.  It is also
2208Perl's exception-trapping mechanism, where the L<C<die>|/die LIST>
2209operator is used to raise exceptions.
2210
2211Before Perl 5.14, the assignment to L<C<$@>|perlvar/$@> occurred before
2212restoration
2213of localized variables, which means that for your code to run on older
2214versions, a temporary is required if you want to mask some, but not all
2215errors:
2216
2217 # alter $@ on nefarious repugnancy only
2218 {
2219    my $e;
2220    {
2221      local $@; # protect existing $@
2222      eval { test_repugnancy() };
2223      # $@ =~ /nefarious/ and die $@; # Perl 5.14 and higher only
2224      $@ =~ /nefarious/ and $e = $@;
2225    }
2226    die $e if defined $e
2227 }
2228
2229There are some different considerations for each form:
2230
2231=over 4
2232
2233=item String eval
2234
2235Since the return value of EXPR is executed as a block within the lexical
2236context of the current Perl program, any outer lexical variables are
2237visible to it, and any package variable settings or subroutine and
2238format definitions remain afterwards.
2239
2240=over 4
2241
2242=item Under the L<C<"unicode_eval"> feature|feature/The 'unicode_eval' and 'evalbytes' features>
2243
2244If this feature is enabled (which is the default under a C<use 5.16> or
2245higher declaration), Perl assumes that EXPR is a character string.
2246Any S<C<use utf8>> or S<C<no utf8>> declarations within
2247the string thus have no effect. Source filters are forbidden as well.
2248(C<unicode_strings>, however, can appear within the string.)
2249
2250See also the L<C<evalbytes>|/evalbytes EXPR> operator, which works properly
2251with source filters.
2252
2253=item Outside the C<"unicode_eval"> feature
2254
2255In this case, the behavior is problematic and is not so easily
2256described.  Here are two bugs that cannot easily be fixed without
2257breaking existing programs:
2258
2259=over 4
2260
2261=item *
2262
2263Perl's internal storage of EXPR affects the behavior of the executed code.
2264For example:
2265
2266    my $v = eval "use utf8; '$expr'";
2267
2268If $expr is C<"\xc4\x80"> (U+0100 in UTF-8), then the value stored in C<$v>
2269will depend on whether Perl stores $expr "upgraded" (cf. L<utf8>) or
2270not:
2271
2272=over
2273
2274=item * If upgraded, C<$v> will be C<"\xc4\x80"> (i.e., the
2275C<use utf8> has no effect.)
2276
2277=item * If non-upgraded, C<$v> will be C<"\x{100}">.
2278
2279=back
2280
2281This is undesirable since being
2282upgraded or not should not affect a string's behavior.
2283
2284=item *
2285
2286Source filters activated within C<eval> leak out into whichever file
2287scope is currently being compiled.  To give an example with the CPAN module
2288L<Semi::Semicolons>:
2289
2290 BEGIN { eval "use Semi::Semicolons; # not filtered" }
2291 # filtered here!
2292
2293L<C<evalbytes>|/evalbytes EXPR> fixes that to work the way one would
2294expect:
2295
2296 use feature "evalbytes";
2297 BEGIN { evalbytes "use Semi::Semicolons; # filtered" }
2298 # not filtered
2299
2300=back
2301
2302=back
2303
2304Problems can arise if the string expands a scalar containing a floating
2305point number.  That scalar can expand to letters, such as C<"NaN"> or
2306C<"Infinity">; or, within the scope of a L<C<use locale>|locale>, the
2307decimal point character may be something other than a dot (such as a
2308comma).  None of these are likely to parse as you are likely expecting.
2309
2310You should be especially careful to remember what's being looked at
2311when:
2312
2313    eval $x;        # CASE 1
2314    eval "$x";      # CASE 2
2315
2316    eval '$x';      # CASE 3
2317    eval { $x };    # CASE 4
2318
2319    eval "\$$x++";  # CASE 5
2320    $$x++;          # CASE 6
2321
2322Cases 1 and 2 above behave identically: they run the code contained in
2323the variable $x.  (Although case 2 has misleading double quotes making
2324the reader wonder what else might be happening (nothing is).)  Cases 3
2325and 4 likewise behave in the same way: they run the code C<'$x'>, which
2326does nothing but return the value of $x.  (Case 4 is preferred for
2327purely visual reasons, but it also has the advantage of compiling at
2328compile-time instead of at run-time.)  Case 5 is a place where
2329normally you I<would> like to use double quotes, except that in this
2330particular situation, you can just use symbolic references instead, as
2331in case 6.
2332
2333An C<eval ''> executed within a subroutine defined
2334in the C<DB> package doesn't see the usual
2335surrounding lexical scope, but rather the scope of the first non-DB piece
2336of code that called it.  You don't normally need to worry about this unless
2337you are writing a Perl debugger.
2338
2339The final semicolon, if any, may be omitted from the value of EXPR.
2340
2341=item Block eval
2342
2343If the code to be executed doesn't vary, you may use the eval-BLOCK
2344form to trap run-time errors without incurring the penalty of
2345recompiling each time.  The error, if any, is still returned in
2346L<C<$@>|perlvar/$@>.
2347Examples:
2348
2349    # make divide-by-zero nonfatal
2350    eval { $answer = $a / $b; }; warn $@ if $@;
2351
2352    # same thing, but less efficient
2353    eval '$answer = $a / $b'; warn $@ if $@;
2354
2355    # a compile-time error
2356    eval { $answer = }; # WRONG
2357
2358    # a run-time error
2359    eval '$answer =';   # sets $@
2360
2361If you want to trap errors when loading an XS module, some problems with
2362the binary interface (such as Perl version skew) may be fatal even with
2363C<eval> unless C<$ENV{PERL_DL_NONLAZY}> is set.  See
2364L<perlrun|perlrun/PERL_DL_NONLAZY>.
2365
2366Using the C<eval {}> form as an exception trap in libraries does have some
2367issues.  Due to the current arguably broken state of C<__DIE__> hooks, you
2368may wish not to trigger any C<__DIE__> hooks that user code may have installed.
2369You can use the C<local $SIG{__DIE__}> construct for this purpose,
2370as this example shows:
2371
2372    # a private exception trap for divide-by-zero
2373    eval { local $SIG{'__DIE__'}; $answer = $a / $b; };
2374    warn $@ if $@;
2375
2376This is especially significant, given that C<__DIE__> hooks can call
2377L<C<die>|/die LIST> again, which has the effect of changing their error
2378messages:
2379
2380    # __DIE__ hooks may modify error messages
2381    {
2382       local $SIG{'__DIE__'} =
2383              sub { (my $x = $_[0]) =~ s/foo/bar/g; die $x };
2384       eval { die "foo lives here" };
2385       print $@ if $@;                # prints "bar lives here"
2386    }
2387
2388Because this promotes action at a distance, this counterintuitive behavior
2389may be fixed in a future release.
2390
2391C<eval BLOCK> does I<not> count as a loop, so the loop control statements
2392L<C<next>|/next LABEL>, L<C<last>|/last LABEL>, or
2393L<C<redo>|/redo LABEL> cannot be used to leave or restart the block.
2394
2395The final semicolon, if any, may be omitted from within the BLOCK.
2396
2397=back
2398
2399=item evalbytes EXPR
2400X<evalbytes>
2401
2402=item evalbytes
2403
2404=for Pod::Functions +evalbytes similar to string eval, but intend to parse a bytestream
2405
2406This function is similar to a L<string eval|/eval EXPR>, except it
2407always parses its argument (or L<C<$_>|perlvar/$_> if EXPR is omitted)
2408as a byte string. If the string contains any code points above 255, then
2409it cannot be a byte string, and the C<evalbytes> will fail with the error
2410stored in C<$@>.
2411
2412C<use utf8> and C<no utf8> within the string have their usual effect.
2413
2414Source filters activated within the evaluated code apply to the code
2415itself.
2416
2417L<C<evalbytes>|/evalbytes EXPR> is available starting in Perl v5.16.  To
2418access it, you must say C<CORE::evalbytes>, but you can omit the
2419C<CORE::> if the
2420L<C<"evalbytes"> feature|feature/The 'unicode_eval' and 'evalbytes' features>
2421is enabled.  This is enabled automatically with a C<use v5.16> (or
2422higher) declaration in the current scope.
2423
2424=item exec LIST
2425X<exec> X<execute>
2426
2427=item exec PROGRAM LIST
2428
2429=for Pod::Functions abandon this program to run another
2430
2431The L<C<exec>|/exec LIST> function executes a system command I<and never
2432returns>; use L<C<system>|/system LIST> instead of L<C<exec>|/exec LIST>
2433if you want it to return.  It fails and
2434returns false only if the command does not exist I<and> it is executed
2435directly instead of via your system's command shell (see below).
2436
2437Since it's a common mistake to use L<C<exec>|/exec LIST> instead of
2438L<C<system>|/system LIST>, Perl warns you if L<C<exec>|/exec LIST> is
2439called in void context and if there is a following statement that isn't
2440L<C<die>|/die LIST>, L<C<warn>|/warn LIST>, or L<C<exit>|/exit EXPR> (if
2441L<warnings> are enabled--but you always do that, right?).  If you
2442I<really> want to follow an L<C<exec>|/exec LIST> with some other
2443statement, you can use one of these styles to avoid the warning:
2444
2445    exec ('foo')   or print STDERR "couldn't exec foo: $!";
2446    { exec ('foo') }; print STDERR "couldn't exec foo: $!";
2447
2448If there is more than one argument in LIST, this calls L<execvp(3)> with the
2449arguments in LIST.  If there is only one element in LIST, the argument is
2450checked for shell metacharacters, and if there are any, the entire
2451argument is passed to the system's command shell for parsing (this is
2452C</bin/sh -c> on Unix platforms, but varies on other platforms).  If
2453there are no shell metacharacters in the argument, it is split into words
2454and passed directly to C<execvp>, which is more efficient.  Examples:
2455
2456    exec '/bin/echo', 'Your arguments are: ', @ARGV;
2457    exec "sort $outfile | uniq";
2458
2459If you don't really want to execute the first argument, but want to lie
2460to the program you are executing about its own name, you can specify
2461the program you actually want to run as an "indirect object" (without a
2462comma) in front of the LIST, as in C<exec PROGRAM LIST>.  (This always
2463forces interpretation of the LIST as a multivalued list, even if there
2464is only a single scalar in the list.)  Example:
2465
2466    my $shell = '/bin/csh';
2467    exec $shell '-sh';    # pretend it's a login shell
2468
2469or, more directly,
2470
2471    exec {'/bin/csh'} '-sh';  # pretend it's a login shell
2472
2473When the arguments get executed via the system shell, results are
2474subject to its quirks and capabilities.  See L<perlop/"`STRING`">
2475for details.
2476
2477Using an indirect object with L<C<exec>|/exec LIST> or
2478L<C<system>|/system LIST> is also more secure.  This usage (which also
2479works fine with L<C<system>|/system LIST>) forces
2480interpretation of the arguments as a multivalued list, even if the
2481list had just one argument.  That way you're safe from the shell
2482expanding wildcards or splitting up words with whitespace in them.
2483
2484    my @args = ( "echo surprise" );
2485
2486    exec @args;               # subject to shell escapes
2487                                # if @args == 1
2488    exec { $args[0] } @args;  # safe even with one-arg list
2489
2490The first version, the one without the indirect object, ran the I<echo>
2491program, passing it C<"surprise"> an argument.  The second version didn't;
2492it tried to run a program named I<"echo surprise">, didn't find it, and set
2493L<C<$?>|perlvar/$?> to a non-zero value indicating failure.
2494
2495On Windows, only the C<exec PROGRAM LIST> indirect object syntax will
2496reliably avoid using the shell; C<exec LIST>, even with more than one
2497element, will fall back to the shell if the first spawn fails.
2498
2499Perl attempts to flush all files opened for output before the exec,
2500but this may not be supported on some platforms (see L<perlport>).
2501To be safe, you may need to set L<C<$E<verbar>>|perlvar/$E<verbar>>
2502(C<$AUTOFLUSH> in L<English>) or call the C<autoflush> method of
2503L<C<IO::Handle>|IO::Handle/METHODS> on any open handles to avoid lost
2504output.
2505
2506Note that L<C<exec>|/exec LIST> will not call your C<END> blocks, nor
2507will it invoke C<DESTROY> methods on your objects.
2508
2509Portability issues: L<perlport/exec>.
2510
2511=item exists EXPR
2512X<exists> X<autovivification>
2513
2514=for Pod::Functions test whether a hash key is present
2515
2516Given an expression that specifies an element of a hash, returns true if the
2517specified element in the hash has ever been initialized, even if the
2518corresponding value is undefined.
2519
2520    print "Exists\n"    if exists $hash{$key};
2521    print "Defined\n"   if defined $hash{$key};
2522    print "True\n"      if $hash{$key};
2523
2524exists may also be called on array elements, but its behavior is much less
2525obvious and is strongly tied to the use of L<C<delete>|/delete EXPR> on
2526arrays.
2527
2528B<WARNING:> Calling L<C<exists>|/exists EXPR> on array values is
2529strongly discouraged.  The
2530notion of deleting or checking the existence of Perl array elements is not
2531conceptually coherent, and can lead to surprising behavior.
2532
2533    print "Exists\n"    if exists $array[$index];
2534    print "Defined\n"   if defined $array[$index];
2535    print "True\n"      if $array[$index];
2536
2537A hash or array element can be true only if it's defined and defined only if
2538it exists, but the reverse doesn't necessarily hold true.
2539
2540Given an expression that specifies the name of a subroutine,
2541returns true if the specified subroutine has ever been declared, even
2542if it is undefined.  Mentioning a subroutine name for exists or defined
2543does not count as declaring it.  Note that a subroutine that does not
2544exist may still be callable: its package may have an C<AUTOLOAD>
2545method that makes it spring into existence the first time that it is
2546called; see L<perlsub>.
2547
2548    print "Exists\n"  if exists &subroutine;
2549    print "Defined\n" if defined &subroutine;
2550
2551Note that the EXPR can be arbitrarily complicated as long as the final
2552operation is a hash or array key lookup or subroutine name:
2553
2554    if (exists $ref->{A}->{B}->{$key})  { }
2555    if (exists $hash{A}{B}{$key})       { }
2556
2557    if (exists $ref->{A}->{B}->[$ix])   { }
2558    if (exists $hash{A}{B}[$ix])        { }
2559
2560    if (exists &{$ref->{A}{B}{$key}})   { }
2561
2562Although the most deeply nested array or hash element will not spring into
2563existence just because its existence was tested, any intervening ones will.
2564Thus C<< $ref->{"A"} >> and C<< $ref->{"A"}->{"B"} >> will spring
2565into existence due to the existence test for the C<$key> element above.
2566This happens anywhere the arrow operator is used, including even here:
2567
2568    undef $ref;
2569    if (exists $ref->{"Some key"})    { }
2570    print $ref;  # prints HASH(0x80d3d5c)
2571
2572Use of a subroutine call, rather than a subroutine name, as an argument
2573to L<C<exists>|/exists EXPR> is an error.
2574
2575    exists &sub;    # OK
2576    exists &sub();  # Error
2577
2578=item exit EXPR
2579X<exit> X<terminate> X<abort>
2580
2581=item exit
2582
2583=for Pod::Functions terminate this program
2584
2585Evaluates EXPR and exits immediately with that value.    Example:
2586
2587    my $ans = <STDIN>;
2588    exit 0 if $ans =~ /^[Xx]/;
2589
2590See also L<C<die>|/die LIST>.  If EXPR is omitted, exits with C<0>
2591status.  The only
2592universally recognized values for EXPR are C<0> for success and C<1>
2593for error; other values are subject to interpretation depending on the
2594environment in which the Perl program is running.  For example, exiting
259569 (EX_UNAVAILABLE) from a I<sendmail> incoming-mail filter will cause
2596the mailer to return the item undelivered, but that's not true everywhere.
2597
2598Don't use L<C<exit>|/exit EXPR> to abort a subroutine if there's any
2599chance that someone might want to trap whatever error happened.  Use
2600L<C<die>|/die LIST> instead, which can be trapped by an
2601L<C<eval>|/eval EXPR>.
2602
2603The L<C<exit>|/exit EXPR> function does not always exit immediately.  It
2604calls any defined C<END> routines first, but these C<END> routines may
2605not themselves abort the exit.  Likewise any object destructors that
2606need to be called are called before the real exit.  C<END> routines and
2607destructors can change the exit status by modifying L<C<$?>|perlvar/$?>.
2608If this is a problem, you can call
2609L<C<POSIX::_exit($status)>|POSIX/C<_exit>> to avoid C<END> and destructor
2610processing.  See L<perlmod> for details.
2611
2612Portability issues: L<perlport/exit>.
2613
2614=item exp EXPR
2615X<exp> X<exponential> X<antilog> X<antilogarithm> X<e>
2616
2617=item exp
2618
2619=for Pod::Functions raise I<e> to a power
2620
2621Returns I<e> (the natural logarithm base) to the power of EXPR.
2622If EXPR is omitted, gives C<exp($_)>.
2623
2624=item fc EXPR
2625X<fc> X<foldcase> X<casefold> X<fold-case> X<case-fold>
2626
2627=item fc
2628
2629=for Pod::Functions +fc return casefolded version of a string
2630
2631Returns the casefolded version of EXPR.  This is the internal function
2632implementing the C<\F> escape in double-quoted strings.
2633
2634Casefolding is the process of mapping strings to a form where case
2635differences are erased; comparing two strings in their casefolded
2636form is effectively a way of asking if two strings are equal,
2637regardless of case.
2638
2639Roughly, if you ever found yourself writing this
2640
2641    lc($this) eq lc($that)    # Wrong!
2642        # or
2643    uc($this) eq uc($that)    # Also wrong!
2644        # or
2645    $this =~ /^\Q$that\E\z/i  # Right!
2646
2647Now you can write
2648
2649    fc($this) eq fc($that)
2650
2651And get the correct results.
2652
2653Perl only implements the full form of casefolding, but you can access
2654the simple folds using L<Unicode::UCD/B<casefold()>> and
2655L<Unicode::UCD/B<prop_invmap()>>.
2656For further information on casefolding, refer to
2657the Unicode Standard, specifically sections 3.13 C<Default Case Operations>,
26584.2 C<Case-Normative>, and 5.18 C<Case Mappings>,
2659available at L<https://www.unicode.org/versions/latest/>, as well as the
2660Case Charts available at L<https://www.unicode.org/charts/case/>.
2661
2662If EXPR is omitted, uses L<C<$_>|perlvar/$_>.
2663
2664This function behaves the same way under various pragmas, such as within
2665L<S<C<"use feature 'unicode_strings">>|feature/The 'unicode_strings' feature>,
2666as L<C<lc>|/lc EXPR> does, with the single exception of
2667L<C<fc>|/fc EXPR> of I<LATIN CAPITAL LETTER SHARP S> (U+1E9E) within the
2668scope of L<S<C<use locale>>|locale>.  The foldcase of this character
2669would normally be C<"ss">, but as explained in the L<C<lc>|/lc EXPR>
2670section, case
2671changes that cross the 255/256 boundary are problematic under locales,
2672and are hence prohibited.  Therefore, this function under locale returns
2673instead the string C<"\x{17F}\x{17F}">, which is the I<LATIN SMALL LETTER
2674LONG S>.  Since that character itself folds to C<"s">, the string of two
2675of them together should be equivalent to a single U+1E9E when foldcased.
2676
2677While the Unicode Standard defines two additional forms of casefolding,
2678one for Turkic languages and one that never maps one character into multiple
2679characters, these are not provided by the Perl core.  However, the CPAN module
2680L<C<Unicode::Casing>|Unicode::Casing> may be used to provide an implementation.
2681
2682L<C<fc>|/fc EXPR> is available only if the
2683L<C<"fc"> feature|feature/The 'fc' feature> is enabled or if it is
2684prefixed with C<CORE::>.  The
2685L<C<"fc"> feature|feature/The 'fc' feature> is enabled automatically
2686with a C<use v5.16> (or higher) declaration in the current scope.
2687
2688=item fcntl FILEHANDLE,FUNCTION,SCALAR
2689X<fcntl>
2690
2691=for Pod::Functions file control system call
2692
2693Implements the L<fcntl(2)> function.  You'll probably have to say
2694
2695    use Fcntl;
2696
2697first to get the correct constant definitions.  Argument processing and
2698value returned work just like L<C<ioctl>|/ioctl
2699FILEHANDLE,FUNCTION,SCALAR> below.  For example:
2700
2701    use Fcntl;
2702    my $flags = fcntl($filehandle, F_GETFL, 0)
2703        or die "Can't fcntl F_GETFL: $!";
2704
2705You don't have to check for L<C<defined>|/defined EXPR> on the return
2706from L<C<fcntl>|/fcntl FILEHANDLE,FUNCTION,SCALAR>.  Like
2707L<C<ioctl>|/ioctl FILEHANDLE,FUNCTION,SCALAR>, it maps a C<0> return
2708from the system call into C<"0 but true"> in Perl.  This string is true
2709in boolean context and C<0> in numeric context.  It is also exempt from
2710the normal
2711L<C<Argument "..." isn't numeric>|perldiag/Argument "%s" isn't numeric%s>
2712L<warnings> on improper numeric conversions.
2713
2714Note that L<C<fcntl>|/fcntl FILEHANDLE,FUNCTION,SCALAR> raises an
2715exception if used on a machine that doesn't implement L<fcntl(2)>.  See
2716the L<Fcntl> module or your L<fcntl(2)> manpage to learn what functions
2717are available on your system.
2718
2719Here's an example of setting a filehandle named C<$REMOTE> to be
2720non-blocking at the system level.  You'll have to negotiate
2721L<C<$E<verbar>>|perlvar/$E<verbar>> on your own, though.
2722
2723    use Fcntl qw(F_GETFL F_SETFL O_NONBLOCK);
2724
2725    my $flags = fcntl($REMOTE, F_GETFL, 0)
2726        or die "Can't get flags for the socket: $!\n";
2727
2728    fcntl($REMOTE, F_SETFL, $flags | O_NONBLOCK)
2729        or die "Can't set flags for the socket: $!\n";
2730
2731Portability issues: L<perlport/fcntl>.
2732
2733=item __FILE__
2734X<__FILE__>
2735
2736=for Pod::Functions the name of the current source file
2737
2738A special token that returns the name of the file in which it occurs.
2739It can be altered by the mechanism described at
2740L<perlsyn/"Plain Old Comments (Not!)">.
2741
2742=item fileno FILEHANDLE
2743X<fileno>
2744
2745=item fileno DIRHANDLE
2746
2747=for Pod::Functions return file descriptor from filehandle
2748
2749Returns the file descriptor for a filehandle or directory handle,
2750or undefined if the
2751filehandle is not open.  If there is no real file descriptor at the OS
2752level, as can happen with filehandles connected to memory objects via
2753L<C<open>|/open FILEHANDLE,MODE,EXPR> with a reference for the third
2754argument, -1 is returned.
2755
2756This is mainly useful for constructing bitmaps for
2757L<C<select>|/select RBITS,WBITS,EBITS,TIMEOUT> and low-level POSIX
2758tty-handling operations.
2759If FILEHANDLE is an expression, the value is taken as an indirect
2760filehandle, generally its name.
2761
2762You can use this to find out whether two handles refer to the
2763same underlying descriptor:
2764
2765    if (fileno($this) != -1 && fileno($this) == fileno($that)) {
2766        print "\$this and \$that are dups\n";
2767    } elsif (fileno($this) != -1 && fileno($that) != -1) {
2768        print "\$this and \$that have different " .
2769            "underlying file descriptors\n";
2770    } else {
2771        print "At least one of \$this and \$that does " .
2772            "not have a real file descriptor\n";
2773    }
2774
2775The behavior of L<C<fileno>|/fileno FILEHANDLE> on a directory handle
2776depends on the operating system.  On a system with L<dirfd(3)> or
2777similar, L<C<fileno>|/fileno FILEHANDLE> on a directory
2778handle returns the underlying file descriptor associated with the
2779handle; on systems with no such support, it returns the undefined value,
2780and sets L<C<$!>|perlvar/$!> (errno).
2781
2782=item flock FILEHANDLE,OPERATION
2783X<flock> X<lock> X<locking>
2784
2785=for Pod::Functions lock an entire file with an advisory lock
2786
2787Calls L<flock(2)>, or an emulation of it, on FILEHANDLE.  Returns true
2788for success, false on failure.  Produces a fatal error if used on a
2789machine that doesn't implement L<flock(2)>, L<fcntl(2)> locking, or
2790L<lockf(3)>.  L<C<flock>|/flock FILEHANDLE,OPERATION> is Perl's portable
2791file-locking interface, although it locks entire files only, not
2792records.
2793
2794Two potentially non-obvious but traditional L<C<flock>|/flock
2795FILEHANDLE,OPERATION> semantics are
2796that it waits indefinitely until the lock is granted, and that its locks
2797are B<merely advisory>.  Such discretionary locks are more flexible, but
2798offer fewer guarantees.  This means that programs that do not also use
2799L<C<flock>|/flock FILEHANDLE,OPERATION> may modify files locked with
2800L<C<flock>|/flock FILEHANDLE,OPERATION>.  See L<perlport>,
2801your port's specific documentation, and your system-specific local manpages
2802for details.  It's best to assume traditional behavior if you're writing
2803portable programs.  (But if you're not, you should as always feel perfectly
2804free to write for your own system's idiosyncrasies (sometimes called
2805"features").  Slavish adherence to portability concerns shouldn't get
2806in the way of your getting your job done.)
2807
2808OPERATION is one of LOCK_SH, LOCK_EX, or LOCK_UN, possibly combined with
2809LOCK_NB.  These constants are traditionally valued 1, 2, 8 and 4, but
2810you can use the symbolic names if you import them from the L<Fcntl> module,
2811either individually, or as a group using the C<:flock> tag.  LOCK_SH
2812requests a shared lock, LOCK_EX requests an exclusive lock, and LOCK_UN
2813releases a previously requested lock.  If LOCK_NB is bitwise-or'ed with
2814LOCK_SH or LOCK_EX, then L<C<flock>|/flock FILEHANDLE,OPERATION> returns
2815immediately rather than blocking waiting for the lock; check the return
2816status to see if you got it.
2817
2818To avoid the possibility of miscoordination, Perl now flushes FILEHANDLE
2819before locking or unlocking it.
2820
2821Note that the emulation built with L<lockf(3)> doesn't provide shared
2822locks, and it requires that FILEHANDLE be open with write intent.  These
2823are the semantics that L<lockf(3)> implements.  Most if not all systems
2824implement L<lockf(3)> in terms of L<fcntl(2)> locking, though, so the
2825differing semantics shouldn't bite too many people.
2826
2827Note that the L<fcntl(2)> emulation of L<flock(3)> requires that FILEHANDLE
2828be open with read intent to use LOCK_SH and requires that it be open
2829with write intent to use LOCK_EX.
2830
2831Note also that some versions of L<C<flock>|/flock FILEHANDLE,OPERATION>
2832cannot lock things over the network; you would need to use the more
2833system-specific L<C<fcntl>|/fcntl FILEHANDLE,FUNCTION,SCALAR> for
2834that.  If you like you can force Perl to ignore your system's L<flock(2)>
2835function, and so provide its own L<fcntl(2)>-based emulation, by passing
2836the switch C<-Ud_flock> to the F<Configure> program when you configure
2837and build a new Perl.
2838
2839Here's a mailbox appender for BSD systems.
2840
2841    # import LOCK_* and SEEK_END constants
2842    use Fcntl qw(:flock SEEK_END);
2843
2844    sub lock {
2845        my ($fh) = @_;
2846        flock($fh, LOCK_EX) or die "Cannot lock mailbox - $!\n";
2847        # and, in case we're running on a very old UNIX
2848        # variant without the modern O_APPEND semantics...
2849        seek($fh, 0, SEEK_END) or die "Cannot seek - $!\n";
2850    }
2851
2852    sub unlock {
2853        my ($fh) = @_;
2854        flock($fh, LOCK_UN) or die "Cannot unlock mailbox - $!\n";
2855    }
2856
2857    open(my $mbox, ">>", "/usr/spool/mail/$ENV{'USER'}")
2858        or die "Can't open mailbox: $!";
2859
2860    lock($mbox);
2861    print $mbox $msg,"\n\n";
2862    unlock($mbox);
2863
2864On systems that support a real L<flock(2)>, locks are inherited across
2865L<C<fork>|/fork> calls, whereas those that must resort to the more
2866capricious L<fcntl(2)> function lose their locks, making it seriously
2867harder to write servers.
2868
2869See also L<DB_File> for other L<C<flock>|/flock FILEHANDLE,OPERATION>
2870examples.
2871
2872Portability issues: L<perlport/flock>.
2873
2874=item fork
2875X<fork> X<child> X<parent>
2876
2877=for Pod::Functions create a new process just like this one
2878
2879Does a L<fork(2)> system call to create a new process running the
2880same program at the same point.  It returns the child pid to the
2881parent process, C<0> to the child process, or L<C<undef>|/undef EXPR> if
2882the fork is
2883unsuccessful.  File descriptors (and sometimes locks on those descriptors)
2884are shared, while everything else is copied.  On most systems supporting
2885L<fork(2)>, great care has gone into making it extremely efficient (for
2886example, using copy-on-write technology on data pages), making it the
2887dominant paradigm for multitasking over the last few decades.
2888
2889Perl attempts to flush all files opened for output before forking the
2890child process, but this may not be supported on some platforms (see
2891L<perlport>).  To be safe, you may need to set
2892L<C<$E<verbar>>|perlvar/$E<verbar>> (C<$AUTOFLUSH> in L<English>) or
2893call the C<autoflush> method of L<C<IO::Handle>|IO::Handle/METHODS> on
2894any open handles to avoid duplicate output.
2895
2896If you L<C<fork>|/fork> without ever waiting on your children, you will
2897accumulate zombies.  On some systems, you can avoid this by setting
2898L<C<$SIG{CHLD}>|perlvar/%SIG> to C<"IGNORE">.  See also L<perlipc> for
2899more examples of forking and reaping moribund children.
2900
2901Note that if your forked child inherits system file descriptors like
2902STDIN and STDOUT that are actually connected by a pipe or socket, even
2903if you exit, then the remote server (such as, say, a CGI script or a
2904backgrounded job launched from a remote shell) won't think you're done.
2905You should reopen those to F</dev/null> if it's any issue.
2906
2907On some platforms such as Windows, where the L<fork(2)> system call is
2908not available, Perl can be built to emulate L<C<fork>|/fork> in the Perl
2909interpreter.  The emulation is designed, at the level of the Perl
2910program, to be as compatible as possible with the "Unix" L<fork(2)>.
2911However it has limitations that have to be considered in code intended
2912to be portable.  See L<perlfork> for more details.
2913
2914Portability issues: L<perlport/fork>.
2915
2916=item format
2917X<format>
2918
2919=for Pod::Functions declare a picture format with use by the write() function
2920
2921Declare a picture format for use by the L<C<write>|/write FILEHANDLE>
2922function.  For example:
2923
2924    format Something =
2925        Test: @<<<<<<<< @||||| @>>>>>
2926              $str,     $%,    '$' . int($num)
2927    .
2928
2929    $str = "widget";
2930    $num = $cost/$quantity;
2931    $~ = 'Something';
2932    write;
2933
2934See L<perlform> for many details and examples.
2935
2936=item formline PICTURE,LIST
2937X<formline>
2938
2939=for Pod::Functions internal function used for formats
2940
2941This is an internal function used by L<C<format>|/format>s, though you
2942may call it, too.  It formats (see L<perlform>) a list of values
2943according to the contents of PICTURE, placing the output into the format
2944output accumulator, L<C<$^A>|perlvar/$^A> (or C<$ACCUMULATOR> in
2945L<English>).  Eventually, when a L<C<write>|/write FILEHANDLE> is done,
2946the contents of L<C<$^A>|perlvar/$^A> are written to some filehandle.
2947You could also read L<C<$^A>|perlvar/$^A> and then set
2948L<C<$^A>|perlvar/$^A> back to C<"">.  Note that a format typically does
2949one L<C<formline>|/formline PICTURE,LIST> per line of form, but the
2950L<C<formline>|/formline PICTURE,LIST> function itself doesn't care how
2951many newlines are embedded in the PICTURE.  This means that the C<~> and
2952C<~~> tokens treat the entire PICTURE as a single line.  You may
2953therefore need to use multiple formlines to implement a single record
2954format, just like the L<C<format>|/format> compiler.
2955
2956Be careful if you put double quotes around the picture, because an C<@>
2957character may be taken to mean the beginning of an array name.
2958L<C<formline>|/formline PICTURE,LIST> always returns true.  See
2959L<perlform> for other examples.
2960
2961If you are trying to use this instead of L<C<write>|/write FILEHANDLE>
2962to capture the output, you may find it easier to open a filehandle to a
2963scalar (C<< open my $fh, ">", \$output >>) and write to that instead.
2964
2965=item getc FILEHANDLE
2966X<getc> X<getchar> X<character> X<file, read>
2967
2968=item getc
2969
2970=for Pod::Functions get the next character from the filehandle
2971
2972Returns the next character from the input file attached to FILEHANDLE,
2973or the undefined value at end of file or if there was an error (in
2974the latter case L<C<$!>|perlvar/$!> is set).  If FILEHANDLE is omitted,
2975reads from
2976STDIN.  This is not particularly efficient.  However, it cannot be
2977used by itself to fetch single characters without waiting for the user
2978to hit enter.  For that, try something more like:
2979
2980    if ($BSD_STYLE) {
2981        system "stty cbreak </dev/tty >/dev/tty 2>&1";
2982    }
2983    else {
2984        system "stty", '-icanon', 'eol', "\001";
2985    }
2986
2987    my $key = getc(STDIN);
2988
2989    if ($BSD_STYLE) {
2990        system "stty -cbreak </dev/tty >/dev/tty 2>&1";
2991    }
2992    else {
2993        system 'stty', 'icanon', 'eol', '^@'; # ASCII NUL
2994    }
2995    print "\n";
2996
2997Determination of whether C<$BSD_STYLE> should be set is left as an
2998exercise to the reader.
2999
3000The L<C<POSIX::getattr>|POSIX/C<getattr>> function can do this more
3001portably on systems purporting POSIX compliance.  See also the
3002L<C<Term::ReadKey>|Term::ReadKey> module on CPAN.
3003
3004=item getlogin
3005X<getlogin> X<login>
3006
3007=for Pod::Functions return who logged in at this tty
3008
3009This implements the C library function of the same name, which on most
3010systems returns the current login from F</etc/utmp>, if any.  If it
3011returns the empty string, use L<C<getpwuid>|/getpwuid UID>.
3012
3013    my $login = getlogin || getpwuid($<) || "Kilroy";
3014
3015Do not consider L<C<getlogin>|/getlogin> for authentication: it is not
3016as secure as L<C<getpwuid>|/getpwuid UID>.
3017
3018Portability issues: L<perlport/getlogin>.
3019
3020=item getpeername SOCKET
3021X<getpeername> X<peer>
3022
3023=for Pod::Functions find the other end of a socket connection
3024
3025Returns the packed sockaddr address of the other end of the SOCKET
3026connection.
3027
3028    use Socket;
3029    my $hersockaddr    = getpeername($sock);
3030    my ($port, $iaddr) = sockaddr_in($hersockaddr);
3031    my $herhostname    = gethostbyaddr($iaddr, AF_INET);
3032    my $herstraddr     = inet_ntoa($iaddr);
3033
3034=item getpgrp PID
3035X<getpgrp> X<group>
3036
3037=for Pod::Functions get process group
3038
3039Returns the current process group for the specified PID.  Use
3040a PID of C<0> to get the current process group for the
3041current process.  Will raise an exception if used on a machine that
3042doesn't implement L<getpgrp(2)>.  If PID is omitted, returns the process
3043group of the current process.  Note that the POSIX version of
3044L<C<getpgrp>|/getpgrp PID> does not accept a PID argument, so only
3045C<PID==0> is truly portable.
3046
3047Portability issues: L<perlport/getpgrp>.
3048
3049=item getppid
3050X<getppid> X<parent> X<pid>
3051
3052=for Pod::Functions get parent process ID
3053
3054Returns the process id of the parent process.
3055
3056Note for Linux users: Between v5.8.1 and v5.16.0 Perl would work
3057around non-POSIX thread semantics the minority of Linux systems (and
3058Debian GNU/kFreeBSD systems) that used LinuxThreads, this emulation
3059has since been removed.  See the documentation for L<$$|perlvar/$$> for
3060details.
3061
3062Portability issues: L<perlport/getppid>.
3063
3064=item getpriority WHICH,WHO
3065X<getpriority> X<priority> X<nice>
3066
3067=for Pod::Functions get current nice value
3068
3069Returns the current priority for a process, a process group, or a user.
3070(See L<getpriority(2)>.)  Will raise a fatal exception if used on a
3071machine that doesn't implement L<getpriority(2)>.
3072
3073C<WHICH> can be any of C<PRIO_PROCESS>, C<PRIO_PGRP> or C<PRIO_USER>
3074imported from L<POSIX/RESOURCE CONSTANTS>.
3075
3076Portability issues: L<perlport/getpriority>.
3077
3078=item getpwnam NAME
3079X<getpwnam> X<getgrnam> X<gethostbyname> X<getnetbyname> X<getprotobyname>
3080X<getpwuid> X<getgrgid> X<getservbyname> X<gethostbyaddr> X<getnetbyaddr>
3081X<getprotobynumber> X<getservbyport> X<getpwent> X<getgrent> X<gethostent>
3082X<getnetent> X<getprotoent> X<getservent> X<setpwent> X<setgrent> X<sethostent>
3083X<setnetent> X<setprotoent> X<setservent> X<endpwent> X<endgrent> X<endhostent>
3084X<endnetent> X<endprotoent> X<endservent>
3085
3086=for Pod::Functions get passwd record given user login name
3087
3088=item getgrnam NAME
3089
3090=for Pod::Functions get group record given group name
3091
3092=item gethostbyname NAME
3093
3094=for Pod::Functions get host record given name
3095
3096=item getnetbyname NAME
3097
3098=for Pod::Functions get networks record given name
3099
3100=item getprotobyname NAME
3101
3102=for Pod::Functions get protocol record given name
3103
3104=item getpwuid UID
3105
3106=for Pod::Functions get passwd record given user ID
3107
3108=item getgrgid GID
3109
3110=for Pod::Functions get group record given group user ID
3111
3112=item getservbyname NAME,PROTO
3113
3114=for Pod::Functions get services record given its name
3115
3116=item gethostbyaddr ADDR,ADDRTYPE
3117
3118=for Pod::Functions get host record given its address
3119
3120=item getnetbyaddr ADDR,ADDRTYPE
3121
3122=for Pod::Functions get network record given its address
3123
3124=item getprotobynumber NUMBER
3125
3126=for Pod::Functions get protocol record numeric protocol
3127
3128=item getservbyport PORT,PROTO
3129
3130=for Pod::Functions get services record given numeric port
3131
3132=item getpwent
3133
3134=for Pod::Functions get next passwd record
3135
3136=item getgrent
3137
3138=for Pod::Functions get next group record
3139
3140=item gethostent
3141
3142=for Pod::Functions get next hosts record
3143
3144=item getnetent
3145
3146=for Pod::Functions get next networks record
3147
3148=item getprotoent
3149
3150=for Pod::Functions get next protocols record
3151
3152=item getservent
3153
3154=for Pod::Functions get next services record
3155
3156=item setpwent
3157
3158=for Pod::Functions prepare passwd file for use
3159
3160=item setgrent
3161
3162=for Pod::Functions prepare group file for use
3163
3164=item sethostent STAYOPEN
3165
3166=for Pod::Functions prepare hosts file for use
3167
3168=item setnetent STAYOPEN
3169
3170=for Pod::Functions prepare networks file for use
3171
3172=item setprotoent STAYOPEN
3173
3174=for Pod::Functions prepare protocols file for use
3175
3176=item setservent STAYOPEN
3177
3178=for Pod::Functions prepare services file for use
3179
3180=item endpwent
3181
3182=for Pod::Functions be done using passwd file
3183
3184=item endgrent
3185
3186=for Pod::Functions be done using group file
3187
3188=item endhostent
3189
3190=for Pod::Functions be done using hosts file
3191
3192=item endnetent
3193
3194=for Pod::Functions be done using networks file
3195
3196=item endprotoent
3197
3198=for Pod::Functions be done using protocols file
3199
3200=item endservent
3201
3202=for Pod::Functions be done using services file
3203
3204These routines are the same as their counterparts in the
3205system C library.  In list context, the return values from the
3206various get routines are as follows:
3207
3208 #    0        1          2           3         4
3209 my ( $name,   $passwd,   $gid,       $members  ) = getgr*
3210 my ( $name,   $aliases,  $addrtype,  $net      ) = getnet*
3211 my ( $name,   $aliases,  $port,      $proto    ) = getserv*
3212 my ( $name,   $aliases,  $proto                ) = getproto*
3213 my ( $name,   $aliases,  $addrtype,  $length,  @addrs ) = gethost*
3214 my ( $name,   $passwd,   $uid,       $gid,     $quota,
3215    $comment,  $gcos,     $dir,       $shell,   $expire ) = getpw*
3216 #    5        6          7           8         9
3217
3218(If the entry doesn't exist, the return value is a single meaningless true
3219value.)
3220
3221The exact meaning of the $gcos field varies but it usually contains
3222the real name of the user (as opposed to the login name) and other
3223information pertaining to the user.  Beware, however, that in many
3224system users are able to change this information and therefore it
3225cannot be trusted and therefore the $gcos is tainted (see
3226L<perlsec>).  The $passwd and $shell, user's encrypted password and
3227login shell, are also tainted, for the same reason.
3228
3229In scalar context, you get the name, unless the function was a
3230lookup by name, in which case you get the other thing, whatever it is.
3231(If the entry doesn't exist you get the undefined value.)  For example:
3232
3233    my $uid   = getpwnam($name);
3234    my $name  = getpwuid($num);
3235    my $name  = getpwent();
3236    my $gid   = getgrnam($name);
3237    my $name  = getgrgid($num);
3238    my $name  = getgrent();
3239    # etc.
3240
3241In I<getpw*()> the fields $quota, $comment, and $expire are special
3242in that they are unsupported on many systems.  If the
3243$quota is unsupported, it is an empty scalar.  If it is supported, it
3244usually encodes the disk quota.  If the $comment field is unsupported,
3245it is an empty scalar.  If it is supported it usually encodes some
3246administrative comment about the user.  In some systems the $quota
3247field may be $change or $age, fields that have to do with password
3248aging.  In some systems the $comment field may be $class.  The $expire
3249field, if present, encodes the expiration period of the account or the
3250password.  For the availability and the exact meaning of these fields
3251in your system, please consult L<getpwnam(3)> and your system's
3252F<pwd.h> file.  You can also find out from within Perl what your
3253$quota and $comment fields mean and whether you have the $expire field
3254by using the L<C<Config>|Config> module and the values C<d_pwquota>, C<d_pwage>,
3255C<d_pwchange>, C<d_pwcomment>, and C<d_pwexpire>.  Shadow password
3256files are supported only if your vendor has implemented them in the
3257intuitive fashion that calling the regular C library routines gets the
3258shadow versions if you're running under privilege or if there exists
3259the L<shadow(3)> functions as found in System V (this includes Solaris
3260and Linux).  Those systems that implement a proprietary shadow password
3261facility are unlikely to be supported.
3262
3263The $members value returned by I<getgr*()> is a space-separated list of
3264the login names of the members of the group.
3265
3266For the I<gethost*()> functions, if the C<h_errno> variable is supported in
3267C, it will be returned to you via L<C<$?>|perlvar/$?> if the function
3268call fails.  The
3269C<@addrs> value returned by a successful call is a list of raw
3270addresses returned by the corresponding library call.  In the
3271Internet domain, each address is four bytes long; you can unpack it
3272by saying something like:
3273
3274    my ($w,$x,$y,$z) = unpack('W4',$addr[0]);
3275
3276The Socket library makes this slightly easier:
3277
3278    use Socket;
3279    my $iaddr = inet_aton("127.1"); # or whatever address
3280    my $name  = gethostbyaddr($iaddr, AF_INET);
3281
3282    # or going the other way
3283    my $straddr = inet_ntoa($iaddr);
3284
3285In the opposite way, to resolve a hostname to the IP address
3286you can write this:
3287
3288    use Socket;
3289    my $packed_ip = gethostbyname("www.perl.org");
3290    my $ip_address;
3291    if (defined $packed_ip) {
3292        $ip_address = inet_ntoa($packed_ip);
3293    }
3294
3295Make sure L<C<gethostbyname>|/gethostbyname NAME> is called in SCALAR
3296context and that its return value is checked for definedness.
3297
3298The L<C<getprotobynumber>|/getprotobynumber NUMBER> function, even
3299though it only takes one argument, has the precedence of a list
3300operator, so beware:
3301
3302    getprotobynumber $number eq 'icmp'   # WRONG
3303    getprotobynumber($number eq 'icmp')  # actually means this
3304    getprotobynumber($number) eq 'icmp'  # better this way
3305
3306If you get tired of remembering which element of the return list
3307contains which return value, by-name interfaces are provided in standard
3308modules: L<C<File::stat>|File::stat>, L<C<Net::hostent>|Net::hostent>,
3309L<C<Net::netent>|Net::netent>, L<C<Net::protoent>|Net::protoent>,
3310L<C<Net::servent>|Net::servent>, L<C<Time::gmtime>|Time::gmtime>,
3311L<C<Time::localtime>|Time::localtime>, and
3312L<C<User::grent>|User::grent>.  These override the normal built-ins,
3313supplying versions that return objects with the appropriate names for
3314each field.  For example:
3315
3316   use File::stat;
3317   use User::pwent;
3318   my $is_his = (stat($filename)->uid == pwent($whoever)->uid);
3319
3320Even though it looks as though they're the same method calls (uid),
3321they aren't, because a C<File::stat> object is different from
3322a C<User::pwent> object.
3323
3324Many of these functions are not safe in a multi-threaded environment
3325where more than one thread can be using them.  In particular, functions
3326like C<getpwent()> iterate per-process and not per-thread, so if two
3327threads are simultaneously iterating, neither will get all the records.
3328
3329Some systems have thread-safe versions of some of the functions, such as
3330C<getpwnam_r()> instead of C<getpwnam()>.  There, Perl automatically and
3331invisibly substitutes the thread-safe version, without notice.  This
3332means that code that safely runs on some systems can fail on others that
3333lack the thread-safe versions.
3334
3335Portability issues: L<perlport/getpwnam> to L<perlport/endservent>.
3336
3337=item getsockname SOCKET
3338X<getsockname>
3339
3340=for Pod::Functions retrieve the sockaddr for a given socket
3341
3342Returns the packed sockaddr address of this end of the SOCKET connection,
3343in case you don't know the address because you have several different
3344IPs that the connection might have come in on.
3345
3346    use Socket;
3347    my $mysockaddr = getsockname($sock);
3348    my ($port, $myaddr) = sockaddr_in($mysockaddr);
3349    printf "Connect to %s [%s]\n",
3350       scalar gethostbyaddr($myaddr, AF_INET),
3351       inet_ntoa($myaddr);
3352
3353=item getsockopt SOCKET,LEVEL,OPTNAME
3354X<getsockopt>
3355
3356=for Pod::Functions get socket options on a given socket
3357
3358Queries the option named OPTNAME associated with SOCKET at a given LEVEL.
3359Options may exist at multiple protocol levels depending on the socket
3360type, but at least the uppermost socket level SOL_SOCKET (defined in the
3361L<C<Socket>|Socket> module) will exist.  To query options at another
3362level the protocol number of the appropriate protocol controlling the
3363option should be supplied.  For example, to indicate that an option is
3364to be interpreted by the TCP protocol, LEVEL should be set to the
3365protocol number of TCP, which you can get using
3366L<C<getprotobyname>|/getprotobyname NAME>.
3367
3368The function returns a packed string representing the requested socket
3369option, or L<C<undef>|/undef EXPR> on error, with the reason for the
3370error placed in L<C<$!>|perlvar/$!>.  Just what is in the packed string
3371depends on LEVEL and OPTNAME; consult L<getsockopt(2)> for details.  A
3372common case is that the option is an integer, in which case the result
3373is a packed integer, which you can decode using
3374L<C<unpack>|/unpack TEMPLATE,EXPR> with the C<i> (or C<I>) format.
3375
3376Here's an example to test whether Nagle's algorithm is enabled on a socket:
3377
3378    use Socket qw(:all);
3379
3380    defined(my $tcp = getprotobyname("tcp"))
3381        or die "Could not determine the protocol number for tcp";
3382    # my $tcp = IPPROTO_TCP; # Alternative
3383    my $packed = getsockopt($socket, $tcp, TCP_NODELAY)
3384        or die "getsockopt TCP_NODELAY: $!";
3385    my $nodelay = unpack("I", $packed);
3386    print "Nagle's algorithm is turned ",
3387           $nodelay ? "off\n" : "on\n";
3388
3389Portability issues: L<perlport/getsockopt>.
3390
3391=item glob EXPR
3392X<glob> X<wildcard> X<filename, expansion> X<expand>
3393
3394=item glob
3395
3396=for Pod::Functions expand filenames using wildcards
3397
3398In list context, returns a (possibly empty) list of filename expansions on
3399the value of EXPR such as the Unix shell Bash would do. In
3400scalar context, glob iterates through such filename expansions, returning
3401L<C<undef>|/undef EXPR> when the list is exhausted. If EXPR is omitted,
3402L<C<$_>|perlvar/$_> is used.
3403
3404    # List context
3405    my @txt_files  = glob("*.txt");
3406    my @perl_files = glob("*.pl *.pm");
3407
3408    # Scalar context
3409    while (my $file = glob("*.mp3")) {
3410        # Do stuff
3411    }
3412
3413Glob also supports an alternate syntax using C<< < >> C<< > >> as
3414delimiters. While this syntax is supported, it is recommended that you
3415use C<glob> instead as it is more readable and searchable.
3416
3417    my @txt_files  = <"*.txt">;
3418
3419If you need case insensitive file globbing that can be achieved using the
3420C<:nocase> parameter of the L<C<bsd_glob>|File::Glob/C<bsd_glob>> module.
3421
3422    use File::Glob qw(:globally :nocase);
3423
3424	my @txt = glob("readme*"); # README readme.txt Readme.md
3425
3426Note that L<C<glob>|/glob EXPR> splits its arguments on whitespace and
3427treats
3428each segment as separate pattern.  As such, C<glob("*.c *.h")>
3429matches all files with a F<.c> or F<.h> extension.  The expression
3430C<glob(".* *")> matches all files in the current working directory.
3431If you want to glob filenames that might contain whitespace, you'll
3432have to use extra quotes around the spacey filename to protect it.
3433For example, to glob filenames that have an C<e> followed by a space
3434followed by an C<f>, use one of:
3435
3436    my @spacies = <"*e f*">;
3437    my @spacies = glob('"*e f*"');
3438    my @spacies = glob(q("*e f*"));
3439
3440If you had to get a variable through, you could do this:
3441
3442    my @spacies = glob("'*${var}e f*'");
3443    my @spacies = glob(qq("*${var}e f*"));
3444
3445If non-empty braces are the only wildcard characters used in the
3446L<C<glob>|/glob EXPR>, no filenames are matched, but potentially many
3447strings are returned.  For example, this produces nine strings, one for
3448each pairing of fruits and colors:
3449
3450    my @many = glob("{apple,tomato,cherry}={green,yellow,red}");
3451
3452This operator is implemented using the standard C<File::Glob> extension.
3453See L<C<bsd_glob>|File::Glob/C<bsd_glob>> for details, including
3454L<C<bsd_glob>|File::Glob/C<bsd_glob>>, which does not treat whitespace
3455as a pattern separator.
3456
3457If a C<glob> expression is used as the condition of a C<while> or C<for>
3458loop, then it will be implicitly assigned to C<$_>.  If either a C<glob>
3459expression or an explicit assignment of a C<glob> expression to a scalar
3460is used as a C<while>/C<for> condition, then the condition actually
3461tests for definedness of the expression's value, not for its regular
3462truth value.
3463
3464Internal implemenation details:
3465
3466This is the internal function implementing the C<< <*.c> >> operator,
3467but you can use it directly. The C<< <*.c> >> operator is discussed in
3468more detail in L<perlop/"I/O Operators">.
3469
3470Portability issues: L<perlport/glob>.
3471
3472=item gmtime EXPR
3473X<gmtime> X<UTC> X<Greenwich>
3474
3475=item gmtime
3476
3477=for Pod::Functions convert UNIX time into record or string using Greenwich time
3478
3479Works just like L<C<localtime>|/localtime EXPR>, but the returned values
3480are localized for the standard Greenwich time zone.
3481
3482Note: When called in list context, $isdst, the last value
3483returned by gmtime, is always C<0>.  There is no
3484Daylight Saving Time in GMT.
3485
3486Portability issues: L<perlport/gmtime>.
3487
3488=item goto LABEL
3489X<goto> X<jump> X<jmp>
3490
3491=item goto EXPR
3492
3493=item goto &NAME
3494
3495=for Pod::Functions create spaghetti code
3496
3497The C<goto LABEL> form finds the statement labeled with LABEL and
3498resumes execution there.  It can't be used to get out of a block or
3499subroutine given to L<C<sort>|/sort SUBNAME LIST>.  It can be used to go
3500almost anywhere else within the dynamic scope, including out of
3501subroutines, but it's usually better to use some other construct such as
3502L<C<last>|/last LABEL> or L<C<die>|/die LIST>.  The author of Perl has
3503never felt the need to use this form of L<C<goto>|/goto LABEL> (in Perl,
3504that is; C is another matter).  (The difference is that C does not offer
3505named loops combined with loop control.  Perl does, and this replaces
3506most structured uses of L<C<goto>|/goto LABEL> in other languages.)
3507
3508The C<goto EXPR> form expects to evaluate C<EXPR> to a code reference or
3509a label name.  If it evaluates to a code reference, it will be handled
3510like C<goto &NAME>, below.  This is especially useful for implementing
3511tail recursion via C<goto __SUB__>.
3512
3513If the expression evaluates to a label name, its scope will be resolved
3514dynamically.  This allows for computed L<C<goto>|/goto LABEL>s per
3515FORTRAN, but isn't necessarily recommended if you're optimizing for
3516maintainability:
3517
3518    goto ("FOO", "BAR", "GLARCH")[$i];
3519
3520As shown in this example, C<goto EXPR> is exempt from the "looks like a
3521function" rule.  A pair of parentheses following it does not (necessarily)
3522delimit its argument.  C<goto("NE")."XT"> is equivalent to C<goto NEXT>.
3523Also, unlike most named operators, this has the same precedence as
3524assignment.
3525
3526Use of C<goto LABEL> or C<goto EXPR> to jump into a construct is
3527deprecated and will issue a warning.  Even then, it may not be used to
3528go into any construct that requires initialization, such as a
3529subroutine, a C<foreach> loop, or a C<given>
3530block.  In general, it may not be used to jump into the parameter
3531of a binary or list operator, but it may be used to jump into the
3532I<first> parameter of a binary operator.  (The C<=>
3533assignment operator's "first" operand is its right-hand
3534operand.)  It also can't be used to go into a
3535construct that is optimized away.
3536
3537The C<goto &NAME> form is quite different from the other forms of
3538L<C<goto>|/goto LABEL>.  In fact, it isn't a goto in the normal sense at
3539all, and doesn't have the stigma associated with other gotos.  Instead,
3540it exits the current subroutine (losing any changes set by
3541L<C<local>|/local EXPR>) and immediately calls in its place the named
3542subroutine using the current value of L<C<@_>|perlvar/@_>.  This is used
3543by C<AUTOLOAD> subroutines that wish to load another subroutine and then
3544pretend that the other subroutine had been called in the first place
3545(except that any modifications to L<C<@_>|perlvar/@_> in the current
3546subroutine are propagated to the other subroutine.) After the
3547L<C<goto>|/goto LABEL>, not even L<C<caller>|/caller EXPR> will be able
3548to tell that this routine was called first.
3549
3550NAME needn't be the name of a subroutine; it can be a scalar variable
3551containing a code reference or a block that evaluates to a code
3552reference.
3553
3554=item grep BLOCK LIST
3555X<grep>
3556
3557=item grep EXPR,LIST
3558
3559=for Pod::Functions locate elements in a list test true against a given criterion
3560
3561This is similar in spirit to, but not the same as, L<grep(1)> and its
3562relatives.  In particular, it is not limited to using regular expressions.
3563
3564Evaluates the BLOCK or EXPR for each element of LIST (locally setting
3565L<C<$_>|perlvar/$_> to each element) and returns the list value
3566consisting of those
3567elements for which the expression evaluated to true.  In scalar
3568context, returns the number of times the expression was true.
3569
3570    my @foo = grep(!/^#/, @bar);    # weed out comments
3571
3572or equivalently,
3573
3574    my @foo = grep {!/^#/} @bar;    # weed out comments
3575
3576Note that L<C<$_>|perlvar/$_> is an alias to the list value, so it can
3577be used to
3578modify the elements of the LIST.  While this is useful and supported,
3579it can cause bizarre results if the elements of LIST are not variables.
3580Similarly, grep returns aliases into the original list, much as a for
3581loop's index variable aliases the list elements.  That is, modifying an
3582element of a list returned by grep (for example, in a C<foreach>,
3583L<C<map>|/map BLOCK LIST> or another L<C<grep>|/grep BLOCK LIST>)
3584actually modifies the element in the original list.
3585This is usually something to be avoided when writing clear code.
3586
3587See also L<C<map>|/map BLOCK LIST> for a list composed of the results of
3588the BLOCK or EXPR.
3589
3590=item hex EXPR
3591X<hex> X<hexadecimal>
3592
3593=item hex
3594
3595=for Pod::Functions convert a hexadecimal string to a number
3596
3597Interprets EXPR as a hex string and returns the corresponding numeric value.
3598If EXPR is omitted, uses L<C<$_>|perlvar/$_>.
3599
3600    print hex '0xAf'; # prints '175'
3601    print hex 'aF';   # same
3602    $valid_input =~ /\A(?:0?[xX])?(?:_?[0-9a-fA-F])*\z/
3603
3604A hex string consists of hex digits and an optional C<0x> or C<x> prefix.
3605Each hex digit may be preceded by a single underscore, which will be ignored.
3606Any other character triggers a warning and causes the rest of the string
3607to be ignored (even leading whitespace, unlike L<C<oct>|/oct EXPR>).
3608Only integers can be represented, and integer overflow triggers a warning.
3609
3610To convert strings that might start with any of C<0>, C<0x>, or C<0b>,
3611see L<C<oct>|/oct EXPR>.  To present something as hex, look into
3612L<C<printf>|/printf FILEHANDLE FORMAT, LIST>,
3613L<C<sprintf>|/sprintf FORMAT, LIST>, and
3614L<C<unpack>|/unpack TEMPLATE,EXPR>.
3615
3616=item import LIST
3617X<import>
3618
3619=for Pod::Functions patch a module's namespace into your own
3620
3621There is no builtin L<C<import>|/import LIST> function.  It is just an
3622ordinary method (subroutine) defined (or inherited) by modules that wish
3623to export names to another module.  The
3624L<C<use>|/use Module VERSION LIST> function calls the
3625L<C<import>|/import LIST> method for the package used.  See also
3626L<C<use>|/use Module VERSION LIST>, L<perlmod>, and L<Exporter>.
3627
3628=item index STR,SUBSTR,POSITION
3629X<index> X<indexOf> X<InStr>
3630
3631=item index STR,SUBSTR
3632
3633=for Pod::Functions find a substring within a string
3634
3635The index function searches for one string within another, but without
3636the wildcard-like behavior of a full regular-expression pattern match.
3637It returns the position of the first occurrence of SUBSTR in STR at
3638or after POSITION.  If POSITION is omitted, starts searching from the
3639beginning of the string.  POSITION before the beginning of the string
3640or after its end is treated as if it were the beginning or the end,
3641respectively.  POSITION and the return value are based at zero.
3642If the substring is not found, L<C<index>|/index STR,SUBSTR,POSITION>
3643returns -1.
3644
3645Find characters or strings:
3646
3647    index("Perl is great", "P");     # Returns 0
3648    index("Perl is great", "g");     # Returns 8
3649    index("Perl is great", "great"); # Also returns 8
3650
3651Attempting to find something not there:
3652
3653    index("Perl is great", "Z");     # Returns -1 (not found)
3654
3655Using an offset to find the I<second> occurrence:
3656
3657    index("Perl is great", "e", 5);  # Returns 10
3658
3659=item int EXPR
3660X<int> X<integer> X<truncate> X<trunc> X<floor>
3661
3662=item int
3663
3664=for Pod::Functions get the integer portion of a number
3665
3666Returns the integer portion of EXPR.  If EXPR is omitted, uses
3667L<C<$_>|perlvar/$_>.
3668You should not use this function for rounding: one because it truncates
3669towards C<0>, and two because machine representations of floating-point
3670numbers can sometimes produce counterintuitive results.  For example,
3671C<int(-6.725/0.025)> produces -268 rather than the correct -269; that's
3672because it's really more like -268.99999999999994315658 instead.  Usually,
3673the L<C<sprintf>|/sprintf FORMAT, LIST>,
3674L<C<printf>|/printf FILEHANDLE FORMAT, LIST>, or the
3675L<C<POSIX::floor>|POSIX/C<floor>> and L<C<POSIX::ceil>|POSIX/C<ceil>>
3676functions will serve you better than will L<C<int>|/int EXPR>.
3677
3678=item ioctl FILEHANDLE,FUNCTION,SCALAR
3679X<ioctl>
3680
3681=for Pod::Functions system-dependent device control system call
3682
3683Implements the L<ioctl(2)> function.  You'll probably first have to say
3684
3685    require "sys/ioctl.ph";  # probably in
3686                             # $Config{archlib}/sys/ioctl.ph
3687
3688to get the correct function definitions.  If F<sys/ioctl.ph> doesn't
3689exist or doesn't have the correct definitions you'll have to roll your
3690own, based on your C header files such as F<< <sys/ioctl.h> >>.
3691(There is a Perl script called B<h2ph> that comes with the Perl kit that
3692may help you in this, but it's nontrivial.)  SCALAR will be read and/or
3693written depending on the FUNCTION; a C pointer to the string value of SCALAR
3694will be passed as the third argument of the actual
3695L<C<ioctl>|/ioctl FILEHANDLE,FUNCTION,SCALAR> call.  (If SCALAR
3696has no string value but does have a numeric value, that value will be
3697passed rather than a pointer to the string value.  To guarantee this to be
3698true, add a C<0> to the scalar before using it.)  The
3699L<C<pack>|/pack TEMPLATE,LIST> and L<C<unpack>|/unpack TEMPLATE,EXPR>
3700functions may be needed to manipulate the values of structures used by
3701L<C<ioctl>|/ioctl FILEHANDLE,FUNCTION,SCALAR>.
3702
3703The return value of L<C<ioctl>|/ioctl FILEHANDLE,FUNCTION,SCALAR> (and
3704L<C<fcntl>|/fcntl FILEHANDLE,FUNCTION,SCALAR>) is as follows:
3705
3706    if OS returns:      then Perl returns:
3707        -1               undefined value
3708         0              string "0 but true"
3709    anything else           that number
3710
3711Thus Perl returns true on success and false on failure, yet you can
3712still easily determine the actual value returned by the operating
3713system:
3714
3715    my $retval = ioctl(...) || -1;
3716    printf "System returned %d\n", $retval;
3717
3718The special string C<"0 but true"> is exempt from
3719L<C<Argument "..." isn't numeric>|perldiag/Argument "%s" isn't numeric%s>
3720L<warnings> on improper numeric conversions.
3721
3722Portability issues: L<perlport/ioctl>.
3723
3724=item join EXPR,LIST
3725X<join>
3726
3727=for Pod::Functions join a list into a string using a separator
3728
3729Joins the separate strings of LIST into a single string with fields
3730separated by the value of EXPR, and returns that new string.  Example:
3731
3732   my $rec = join(':', $login,$passwd,$uid,$gid,$gcos,$home,$shell);
3733
3734Beware that unlike L<C<split>|/split E<sol>PATTERNE<sol>,EXPR,LIMIT>,
3735L<C<join>|/join EXPR,LIST> doesn't take a pattern as its first argument.
3736Compare L<C<split>|/split E<sol>PATTERNE<sol>,EXPR,LIMIT>.
3737
3738=item keys HASH
3739X<keys> X<key>
3740
3741=item keys ARRAY
3742
3743=for Pod::Functions retrieve list of indices from a hash
3744
3745Called in list context, returns a list consisting of all the keys of the
3746named hash, or in Perl 5.12 or later only, the indices of an array.  Perl
3747releases prior to 5.12 will produce a syntax error if you try to use an
3748array argument.  In scalar context, returns the number of keys or indices.
3749
3750Hash entries are returned in an apparently random order.  The actual random
3751order is specific to a given hash; the exact same series of operations
3752on two hashes may result in a different order for each hash.  Any insertion
3753into the hash may change the order, as will any deletion, with the exception
3754that the most recent key returned by L<C<each>|/each HASH> or
3755L<C<keys>|/keys HASH> may be deleted without changing the order.  So
3756long as a given hash is unmodified you may rely on
3757L<C<keys>|/keys HASH>, L<C<values>|/values HASH> and L<C<each>|/each
3758HASH> to repeatedly return the same order
3759as each other.  See L<perlsec/"Algorithmic Complexity Attacks"> for
3760details on why hash order is randomized.  Aside from the guarantees
3761provided here the exact details of Perl's hash algorithm and the hash
3762traversal order are subject to change in any release of Perl.  Tied hashes
3763may behave differently to Perl's hashes with respect to changes in order on
3764insertion and deletion of items.
3765
3766As a side effect, calling L<C<keys>|/keys HASH> resets the internal
3767iterator of the HASH or ARRAY (see L<C<each>|/each HASH>) before
3768yielding the keys.  In
3769particular, calling L<C<keys>|/keys HASH> in void context resets the
3770iterator with no other overhead.
3771
3772Here is yet another way to print your environment:
3773
3774    my @keys = keys %ENV;
3775    my @values = values %ENV;
3776    while (@keys) {
3777        print pop(@keys), '=', pop(@values), "\n";
3778    }
3779
3780or how about sorted by key:
3781
3782    foreach my $key (sort(keys %ENV)) {
3783        print $key, '=', $ENV{$key}, "\n";
3784    }
3785
3786The returned values are copies of the original keys in the hash, so
3787modifying them will not affect the original hash.  Compare
3788L<C<values>|/values HASH>.
3789
3790To sort a hash by value, you'll need to use a
3791L<C<sort>|/sort SUBNAME LIST> function.  Here's a descending numeric
3792sort of a hash by its values:
3793
3794    foreach my $key (sort { $hash{$b} <=> $hash{$a} } keys %hash) {
3795        printf "%4d %s\n", $hash{$key}, $key;
3796    }
3797
3798Used as an lvalue, L<C<keys>|/keys HASH> allows you to increase the
3799number of hash buckets
3800allocated for the given hash.  This can gain you a measure of efficiency if
3801you know the hash is going to get big.  (This is similar to pre-extending
3802an array by assigning a larger number to $#array.)  If you say
3803
3804    keys %hash = 200;
3805
3806then C<%hash> will have at least 200 buckets allocated for it--256 of them,
3807in fact, since it rounds up to the next power of two.  These
3808buckets will be retained even if you do C<%hash = ()>, use C<undef
3809%hash> if you want to free the storage while C<%hash> is still in scope.
3810You can't shrink the number of buckets allocated for the hash using
3811L<C<keys>|/keys HASH> in this way (but you needn't worry about doing
3812this by accident, as trying has no effect).  C<keys @array> in an lvalue
3813context is a syntax error.
3814
3815Starting with Perl 5.14, an experimental feature allowed
3816L<C<keys>|/keys HASH> to take a scalar expression. This experiment has
3817been deemed unsuccessful, and was removed as of Perl 5.24.
3818
3819To avoid confusing would-be users of your code who are running earlier
3820versions of Perl with mysterious syntax errors, put this sort of thing at
3821the top of your file to signal that your code will work I<only> on Perls of
3822a recent vintage:
3823
3824    use v5.12;	# so keys/values/each work on arrays
3825
3826See also L<C<each>|/each HASH>, L<C<values>|/values HASH>, and
3827L<C<sort>|/sort SUBNAME LIST>.
3828
3829=item kill SIGNAL, LIST
3830
3831=item kill SIGNAL
3832X<kill> X<signal>
3833
3834=for Pod::Functions send a signal to a process or process group
3835
3836Sends a signal to a list of processes.  Returns the number of arguments
3837that were successfully used to signal (which is not necessarily the same
3838as the number of processes actually killed, e.g. where a process group is
3839killed).
3840
3841    my $cnt = kill 'HUP', $child1, $child2;
3842    kill 'KILL', @goners;
3843
3844SIGNAL may be either a signal name (a string) or a signal number.  A signal
3845name may start with a C<SIG> prefix, thus C<FOO> and C<SIGFOO> refer to the
3846same signal.  The string form of SIGNAL is recommended for portability because
3847the same signal may have different numbers in different operating systems.
3848
3849A list of signal names supported by the current platform can be found in
3850C<$Config{sig_name}>, which is provided by the L<C<Config>|Config>
3851module.  See L<Config> for more details.
3852
3853A negative signal name is the same as a negative signal number, killing process
3854groups instead of processes.  For example, C<kill '-KILL', $pgrp> and
3855C<kill -9, $pgrp> will send C<SIGKILL> to
3856the entire process group specified.  That
3857means you usually want to use positive not negative signals.
3858
3859If SIGNAL is either the number 0 or the string C<ZERO> (or C<SIGZERO>),
3860no signal is sent to the process, but L<C<kill>|/kill SIGNAL, LIST>
3861checks whether it's I<possible> to send a signal to it
3862(that means, to be brief, that the process is owned by the same user, or we are
3863the super-user).  This is useful to check that a child process is still
3864alive (even if only as a zombie) and hasn't changed its UID.  See
3865L<perlport> for notes on the portability of this construct.
3866
3867The behavior of kill when a I<PROCESS> number is zero or negative depends on
3868the operating system.  For example, on POSIX-conforming systems, zero will
3869signal the current process group, -1 will signal all processes, and any
3870other negative PROCESS number will act as a negative signal number and
3871kill the entire process group specified.
3872
3873If both the SIGNAL and the PROCESS are negative, the results are undefined.
3874A warning may be produced in a future version.
3875
3876See L<perlipc/"Signals"> for more details.
3877
3878On some platforms such as Windows where the L<fork(2)> system call is not
3879available, Perl can be built to emulate L<C<fork>|/fork> at the
3880interpreter level.
3881This emulation has limitations related to kill that have to be considered,
3882for code running on Windows and in code intended to be portable.
3883
3884See L<perlfork> for more details.
3885
3886If there is no I<LIST> of processes, no signal is sent, and the return
3887value is 0.  This form is sometimes used, however, because it causes
3888tainting checks to be run, if your perl support taint checks.  But see
3889L<perlsec/Laundering and Detecting Tainted Data>.
3890
3891Portability issues: L<perlport/kill>.
3892
3893=item last LABEL
3894X<last> X<break>
3895
3896=item last EXPR
3897
3898=item last
3899
3900=for Pod::Functions exit a block prematurely
3901
3902The L<C<last>|/last LABEL> command is like the C<break> statement in C
3903(as used in
3904loops); it immediately exits the loop in question.  If the LABEL is
3905omitted, the command refers to the innermost enclosing
3906loop.  The C<last EXPR> form, available starting in Perl
39075.18.0, allows a label name to be computed at run time,
3908and is otherwise identical to C<last LABEL>.  The
3909L<C<continue>|/continue BLOCK> block, if any, is not executed:
3910
3911    LINE: while (<STDIN>) {
3912        last LINE if /^$/;  # exit when done with header
3913        #...
3914    }
3915
3916L<C<last>|/last LABEL> cannot return a value from a block that typically
3917returns a value, such as C<eval {}>, C<sub {}>, or C<do {}>. It will perform
3918its flow control behavior, which precludes any return value. It should not be
3919used to exit a L<C<grep>|/grep BLOCK LIST> or L<C<map>|/map BLOCK LIST>
3920operation.
3921
3922Note that a block by itself is semantically identical to a loop
3923that executes once.  Thus L<C<last>|/last LABEL> can be used to effect
3924an early exit out of such a block.
3925
3926See also L<C<continue>|/continue BLOCK> for an illustration of how
3927L<C<last>|/last LABEL>, L<C<next>|/next LABEL>, and
3928L<C<redo>|/redo LABEL> work.
3929
3930Unlike most named operators, this has the same precedence as assignment.
3931It is also exempt from the looks-like-a-function rule, so
3932C<last ("foo")."bar"> will cause "bar" to be part of the argument to
3933L<C<last>|/last LABEL>.
3934
3935=item lc EXPR
3936X<lc> X<lowercase>
3937
3938=item lc
3939
3940=for Pod::Functions return lower-case version of a string
3941
3942Returns a lowercased version of EXPR.  If EXPR is omitted, uses
3943L<C<$_>|perlvar/$_>.
3944
3945    my $str = lc("Perl is GREAT"); # "perl is great"
3946
3947What gets returned depends on several factors:
3948
3949=over
3950
3951=item If C<use bytes> is in effect:
3952
3953The results follow ASCII rules.  Only the characters C<A-Z> change,
3954to C<a-z> respectively.
3955
3956=item Otherwise, if C<use locale> for C<LC_CTYPE> is in effect:
3957
3958Respects current C<LC_CTYPE> locale for code points < 256; and uses Unicode
3959rules for the remaining code points (this last can only happen if
3960the UTF8 flag is also set).  See L<perllocale>.
3961
3962Starting in v5.20, Perl uses full Unicode rules if the locale is
3963UTF-8.  Otherwise, there is a deficiency in this scheme, which is that
3964case changes that cross the 255/256
3965boundary are not well-defined.  For example, the lower case of LATIN CAPITAL
3966LETTER SHARP S (U+1E9E) in Unicode rules is U+00DF (on ASCII
3967platforms).   But under C<use locale> (prior to v5.20 or not a UTF-8
3968locale), the lower case of U+1E9E is
3969itself, because 0xDF may not be LATIN SMALL LETTER SHARP S in the
3970current locale, and Perl has no way of knowing if that character even
3971exists in the locale, much less what code point it is.  Perl returns
3972a result that is above 255 (almost always the input character unchanged),
3973for all instances (and there aren't many) where the 255/256 boundary
3974would otherwise be crossed; and starting in v5.22, it raises a
3975L<locale|perldiag/Can't do %s("%s") on non-UTF-8 locale; resolved to "%s".> warning.
3976
3977=item Otherwise, If EXPR has the UTF8 flag set:
3978
3979Unicode rules are used for the case change.
3980
3981=item Otherwise, if C<use feature 'unicode_strings'> or C<use locale ':not_characters'> is in effect:
3982
3983Unicode rules are used for the case change.
3984
3985=item Otherwise:
3986
3987ASCII rules are used for the case change.  The lowercase of any character
3988outside the ASCII range is the character itself.
3989
3990=back
3991
3992B<Note:> This is the internal function implementing the
3993L<C<\L>|perlop/"Quote and Quote-like Operators"> escape in double-quoted
3994strings.
3995
3996    my $str = "Perl is \LGREAT\E"; # "Perl is great"
3997
3998=item lcfirst EXPR
3999X<lcfirst> X<lowercase>
4000
4001=item lcfirst
4002
4003=for Pod::Functions return a string with just the next letter in lower case
4004
4005Returns the value of EXPR with the first character lowercased.  This
4006is the internal function implementing the C<\l> escape in
4007double-quoted strings.
4008
4009If EXPR is omitted, uses L<C<$_>|perlvar/$_>.
4010
4011This function behaves the same way under various pragmas, such as in a locale,
4012as L<C<lc>|/lc EXPR> does.
4013
4014=item length EXPR
4015X<length> X<size>
4016
4017=item length
4018
4019=for Pod::Functions return the number of characters in a string
4020
4021Returns the length in I<characters> of the value of EXPR.  If EXPR is
4022omitted, returns the length of L<C<$_>|perlvar/$_>.  If EXPR is
4023undefined, returns L<C<undef>|/undef EXPR>.
4024
4025This function cannot be used on an entire array or hash to find out how
4026many elements these have.  For that, use C<scalar @array> and C<scalar keys
4027%hash>, respectively.
4028
4029Like all Perl character operations, L<C<length>|/length EXPR> normally
4030deals in logical
4031characters, not physical bytes.  For how many bytes a string encoded as
4032UTF-8 would take up, use C<length(Encode::encode('UTF-8', EXPR))>
4033(you'll have to C<use Encode> first).  See L<Encode> and L<perlunicode>.
4034
4035=item __LINE__
4036X<__LINE__>
4037
4038=for Pod::Functions the current source line number
4039
4040A special token that compiles to the current line number.
4041It can be altered by the mechanism described at
4042L<perlsyn/"Plain Old Comments (Not!)">.
4043
4044=item link OLDFILE,NEWFILE
4045X<link>
4046
4047=for Pod::Functions create a hard link in the filesystem
4048
4049Creates a new filename linked to the old filename.  Returns true for
4050success, false otherwise.
4051
4052Portability issues: L<perlport/link>.
4053
4054=item listen SOCKET,QUEUESIZE
4055X<listen>
4056
4057=for Pod::Functions register your socket as a server
4058
4059Does the same thing that the L<listen(2)> system call does.  Returns true if
4060it succeeded, false otherwise.  See the example in
4061L<perlipc/"Sockets: Client/Server Communication">.
4062
4063=item local EXPR
4064X<local>
4065
4066=for Pod::Functions create a temporary value for a global variable (dynamic scoping)
4067
4068You really probably want to be using L<C<my>|/my VARLIST> instead,
4069because L<C<local>|/local EXPR> isn't what most people think of as
4070"local".  See L<perlsub/"Private Variables via my()"> for details.
4071
4072A local modifies the listed variables to be local to the enclosing
4073block, file, or eval.  If more than one value is listed, the list must
4074be placed in parentheses.  See L<perlsub/"Temporary Values via local()">
4075for details, including issues with tied arrays and hashes.
4076
4077The C<delete local EXPR> construct can also be used to localize the deletion
4078of array/hash elements to the current block.
4079See L<perlsub/"Localized deletion of elements of composite types">.
4080
4081=item localtime EXPR
4082X<localtime> X<ctime>
4083
4084=item localtime
4085
4086=for Pod::Functions convert UNIX time into record or string using local time
4087
4088Converts a time as returned by the time function to a 9-element list
4089with the time analyzed for the local time zone.  Typically used as
4090follows:
4091
4092    #     0    1    2     3     4    5     6     7     8
4093    my ($sec,$min,$hour,$mday,$mon,$year,$wday,$yday,$isdst) =
4094                                                localtime(time);
4095
4096All list elements are numeric and come straight out of the C `struct
4097tm'.  C<$sec>, C<$min>, and C<$hour> are the seconds, minutes, and hours
4098of the specified time.
4099
4100C<$mday> is the day of the month and C<$mon> the month in
4101the range C<0..11>, with 0 indicating January and 11 indicating December.
4102This makes it easy to get a month name from a list:
4103
4104    my @abbr = qw(Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec);
4105    print "$abbr[$mon] $mday";
4106    # $mon=9, $mday=18 gives "Oct 18"
4107
4108C<$year> contains the number of years since 1900.  To get a 4-digit
4109year write:
4110
4111    $year += 1900;
4112
4113To get the last two digits of the year (e.g., "01" in 2001) do:
4114
4115    $year = sprintf("%02d", $year % 100);
4116
4117C<$wday> is the day of the week, with 0 indicating Sunday and 3 indicating
4118Wednesday.  C<$yday> is the day of the year, in the range C<0..364>
4119(or C<0..365> in leap years.)
4120
4121C<$isdst> is true if the specified time occurs when Daylight Saving
4122Time is in effect, false otherwise.
4123
4124If EXPR is omitted, L<C<localtime>|/localtime EXPR> uses the current
4125time (as returned by L<C<time>|/time>).
4126
4127In scalar context, L<C<localtime>|/localtime EXPR> returns the
4128L<ctime(3)> value:
4129
4130 my $now_string = localtime;  # e.g., "Thu Oct 13 04:54:34 1994"
4131
4132This scalar value is always in English, and is B<not> locale-dependent.
4133To get similar but locale-dependent date strings, try for example:
4134
4135 use POSIX qw(strftime);
4136 my $now_string = strftime "%a %b %e %H:%M:%S %Y", localtime;
4137 # or for GMT formatted appropriately for your locale:
4138 my $now_string = strftime "%a %b %e %H:%M:%S %Y", gmtime;
4139
4140C$now_string> will be formatted according to the current LC_TIME locale
4141the program or thread is running in.  See L<perllocale> for how to set
4142up and change that locale.  Note that C<%a> and C<%b>, the short forms
4143of the day of the week and the month of the year, may not necessarily be
4144three characters wide.
4145
4146The L<Time::gmtime> and L<Time::localtime> modules provide a convenient,
4147by-name access mechanism to the L<C<gmtime>|/gmtime EXPR> and
4148L<C<localtime>|/localtime EXPR> functions, respectively.
4149
4150For a comprehensive date and time representation look at the
4151L<DateTime> module on CPAN.
4152
4153For GMT instead of local time use the L<C<gmtime>|/gmtime EXPR> builtin.
4154
4155See also the L<C<Time::Local>|Time::Local> module (for converting
4156seconds, minutes, hours, and such back to the integer value returned by
4157L<C<time>|/time>), and the L<POSIX> module's
4158L<C<mktime>|POSIX/C<mktime>> function.
4159
4160Portability issues: L<perlport/localtime>.
4161
4162=item lock THING
4163X<lock>
4164
4165=for Pod::Functions +5.005 get a thread lock on a variable, subroutine, or method
4166
4167This function places an advisory lock on a shared variable or referenced
4168object contained in I<THING> until the lock goes out of scope.
4169
4170The value returned is the scalar itself, if the argument is a scalar, or a
4171reference, if the argument is a hash, array or subroutine.
4172
4173L<C<lock>|/lock THING> is a "weak keyword"; this means that if you've
4174defined a function
4175by this name (before any calls to it), that function will be called
4176instead.  If you are not under C<use threads::shared> this does nothing.
4177See L<threads::shared>.
4178
4179=item log EXPR
4180X<log> X<logarithm> X<e> X<ln> X<base>
4181
4182=item log
4183
4184=for Pod::Functions retrieve the natural logarithm for a number
4185
4186Returns the natural logarithm (base I<e>) of EXPR.  If EXPR is omitted,
4187returns the log of L<C<$_>|perlvar/$_>.  To get the
4188log of another base, use basic algebra:
4189The base-N log of a number is equal to the natural log of that number
4190divided by the natural log of N.  For example:
4191
4192    sub log10 {
4193        my $n = shift;
4194        return log($n)/log(10);
4195    }
4196
4197See also L<C<exp>|/exp EXPR> for the inverse operation.
4198
4199=item lstat FILEHANDLE
4200X<lstat>
4201
4202=item lstat EXPR
4203
4204=item lstat DIRHANDLE
4205
4206=item lstat
4207
4208=for Pod::Functions stat a symbolic link
4209
4210Does the same thing as the L<C<stat>|/stat FILEHANDLE> function
4211(including setting the special C<_> filehandle) but stats a symbolic
4212link instead of the file the symbolic link points to.  If symbolic links
4213are unimplemented on your system, a normal L<C<stat>|/stat FILEHANDLE>
4214is done.  For much more detailed information, please see the
4215documentation for L<C<stat>|/stat FILEHANDLE>.
4216
4217If EXPR is omitted, stats L<C<$_>|perlvar/$_>.
4218
4219Portability issues: L<perlport/lstat>.
4220
4221=item m//
4222
4223=for Pod::Functions match a string with a regular expression pattern
4224
4225The match operator.  See L<perlop/"Regexp Quote-Like Operators">.
4226
4227=item map BLOCK LIST
4228X<map>
4229
4230=item map EXPR,LIST
4231
4232=for Pod::Functions apply a change to a list to get back a new list with the changes
4233
4234Evaluates the BLOCK or EXPR for each element of LIST (locally setting
4235L<C<$_>|perlvar/$_> to each element) and composes a list of the results of
4236each such evaluation.  Each element of LIST may produce zero, one, or more
4237elements in the generated list, so the number of elements in the generated
4238list may differ from that in LIST.  In scalar context, returns the total
4239number of elements so generated.  In list context, returns the generated list.
4240
4241    my @chars = map(chr, @numbers);
4242
4243translates a list of numbers to the corresponding characters.
4244
4245    my @squares = map { $_ * $_ } @numbers;
4246
4247translates a list of numbers to their squared values.
4248
4249    my @squares = map { $_ > 5 ? ($_ * $_) : () } @numbers;
4250
4251shows that number of returned elements can differ from the number of
4252input elements.  To omit an element, return an empty list ().
4253This could also be achieved by writing
4254
4255    my @squares = map { $_ * $_ } grep { $_ > 5 } @numbers;
4256
4257which makes the intention more clear.
4258
4259Map always returns a list, which can be
4260assigned to a hash such that the elements
4261become key/value pairs.  See L<perldata> for more details.
4262
4263    my %hash = map { get_a_key_for($_) => $_ } @array;
4264
4265is just a funny way to write
4266
4267    my %hash;
4268    foreach (@array) {
4269        $hash{get_a_key_for($_)} = $_;
4270    }
4271
4272Note that L<C<$_>|perlvar/$_> is an alias to the list value, so it can
4273be used to modify the elements of the LIST.  While this is useful and
4274supported, it can cause bizarre results if the elements of LIST are not
4275variables.  Using a regular C<foreach> loop for this purpose would be
4276clearer in most cases.  See also L<C<grep>|/grep BLOCK LIST> for a
4277list composed of those items of the original list for which the BLOCK
4278or EXPR evaluates to true.
4279
4280C<{> starts both hash references and blocks, so C<map { ...> could be either
4281the start of map BLOCK LIST or map EXPR, LIST.  Because Perl doesn't look
4282ahead for the closing C<}> it has to take a guess at which it's dealing with
4283based on what it finds just after the
4284C<{>.  Usually it gets it right, but if it
4285doesn't it won't realize something is wrong until it gets to the C<}> and
4286encounters the missing (or unexpected) comma.  The syntax error will be
4287reported close to the C<}>, but you'll need to change something near the C<{>
4288such as using a unary C<+> or semicolon to give Perl some help:
4289
4290 my %hash = map {  "\L$_" => 1  } @array # perl guesses EXPR. wrong
4291 my %hash = map { +"\L$_" => 1  } @array # perl guesses BLOCK. right
4292 my %hash = map {; "\L$_" => 1  } @array # this also works
4293 my %hash = map { ("\L$_" => 1) } @array # as does this
4294 my %hash = map {  lc($_) => 1  } @array # and this.
4295 my %hash = map +( lc($_) => 1 ), @array # this is EXPR and works!
4296
4297 my %hash = map  ( lc($_), 1 ),   @array # evaluates to (1, @array)
4298
4299or to force an anon hash constructor use C<+{>:
4300
4301    my @hashes = map +{ lc($_) => 1 }, @array # EXPR, so needs
4302                                              # comma at end
4303
4304to get a list of anonymous hashes each with only one entry apiece.
4305
4306=item mkdir FILENAME,MODE
4307X<mkdir> X<md> X<directory, create>
4308
4309=item mkdir FILENAME
4310
4311=item mkdir
4312
4313=for Pod::Functions create a directory
4314
4315Creates the directory specified by FILENAME, with permissions
4316specified by MODE (as modified by L<C<umask>|/umask EXPR>).  If it
4317succeeds it returns true; otherwise it returns false and sets
4318L<C<$!>|perlvar/$!> (errno).
4319MODE defaults to 0777 if omitted, and FILENAME defaults
4320to L<C<$_>|perlvar/$_> if omitted.
4321
4322In general, it is better to create directories with a permissive MODE
4323and let the user modify that with their L<C<umask>|/umask EXPR> than it
4324is to supply
4325a restrictive MODE and give the user no way to be more permissive.
4326The exceptions to this rule are when the file or directory should be
4327kept private (mail files, for instance).  The documentation for
4328L<C<umask>|/umask EXPR> discusses the choice of MODE in more detail.
4329
4330Note that according to the POSIX 1003.1-1996 the FILENAME may have any
4331number of trailing slashes.  Some operating and filesystems do not get
4332this right, so Perl automatically removes all trailing slashes to keep
4333everyone happy.
4334
4335To recursively create a directory structure, look at
4336the L<C<make_path>|File::Path/make_path( $dir1, $dir2, .... )> function
4337of the L<File::Path> module.
4338
4339=item msgctl ID,CMD,ARG
4340X<msgctl>
4341
4342=for Pod::Functions SysV IPC message control operations
4343
4344Calls the System V IPC function L<msgctl(2)>.  You'll probably have to say
4345
4346    use IPC::SysV;
4347
4348first to get the correct constant definitions.  If CMD is C<IPC_STAT>,
4349then ARG must be a variable that will hold the returned C<msqid_ds>
4350structure.  Returns like L<C<ioctl>|/ioctl FILEHANDLE,FUNCTION,SCALAR>:
4351the undefined value for error, C<"0 but true"> for zero, or the actual
4352return value otherwise.  See also L<perlipc/"SysV IPC"> and the
4353documentation for L<C<IPC::SysV>|IPC::SysV> and
4354L<C<IPC::Semaphore>|IPC::Semaphore>.
4355
4356Portability issues: L<perlport/msgctl>.
4357
4358=item msgget KEY,FLAGS
4359X<msgget>
4360
4361=for Pod::Functions get SysV IPC message queue
4362
4363Calls the System V IPC function L<msgget(2)>.  Returns the message queue
4364id, or L<C<undef>|/undef EXPR> on error.  See also L<perlipc/"SysV IPC">
4365and the documentation for L<C<IPC::SysV>|IPC::SysV> and
4366L<C<IPC::Msg>|IPC::Msg>.
4367
4368Portability issues: L<perlport/msgget>.
4369
4370=item msgrcv ID,VAR,SIZE,TYPE,FLAGS
4371X<msgrcv>
4372
4373=for Pod::Functions receive a SysV IPC message from a message queue
4374
4375Calls the System V IPC function msgrcv to receive a message from
4376message queue ID into variable VAR with a maximum message size of
4377SIZE.  Note that when a message is received, the message type as a
4378native long integer will be the first thing in VAR, followed by the
4379actual message.  This packing may be opened with C<unpack("l! a*")>.
4380Taints the variable.  Returns true if successful, false
4381on error.  See also L<perlipc/"SysV IPC"> and the documentation for
4382L<C<IPC::SysV>|IPC::SysV> and L<C<IPC::Msg>|IPC::Msg>.
4383
4384Portability issues: L<perlport/msgrcv>.
4385
4386=item msgsnd ID,MSG,FLAGS
4387X<msgsnd>
4388
4389=for Pod::Functions send a SysV IPC message to a message queue
4390
4391Calls the System V IPC function msgsnd to send the message MSG to the
4392message queue ID.  MSG must begin with the native long integer message
4393type, followed by the message itself.  This kind of packing can be achieved
4394with C<pack("l! a*", $type, $message)>.  Returns true if successful,
4395false on error.  See also L<perlipc/"SysV IPC"> and the documentation
4396for L<C<IPC::SysV>|IPC::SysV> and L<C<IPC::Msg>|IPC::Msg>.
4397
4398Portability issues: L<perlport/msgsnd>.
4399
4400=item my VARLIST
4401X<my>
4402
4403=item my TYPE VARLIST
4404
4405=item my VARLIST : ATTRS
4406
4407=item my TYPE VARLIST : ATTRS
4408
4409=for Pod::Functions declare and assign a local variable (lexical scoping)
4410
4411A L<C<my>|/my VARLIST> declares the listed variables to be local
4412(lexically) to the enclosing block, file, or L<C<eval>|/eval EXPR>.  If
4413more than one variable is listed, the list must be placed in
4414parentheses.
4415
4416Note that with a parenthesised list, L<C<undef>|/undef EXPR> can be used
4417as a dummy placeholder, for example to skip assignment of initial
4418values:
4419
4420    my ( undef, $min, $hour ) = localtime;
4421
4422Redeclaring a variable in the same scope or statement will "shadow" the
4423previous declaration, creating a new instance and preventing access to
4424the previous one. This is usually undesired and, if warnings are enabled,
4425will result in a warning in the C<shadow> category.
4426
4427The exact semantics and interface of TYPE and ATTRS are still
4428evolving.  TYPE may be a bareword, a constant declared
4429with L<C<use constant>|constant>, or L<C<__PACKAGE__>|/__PACKAGE__>.  It
4430is
4431currently bound to the use of the L<fields> pragma,
4432and attributes are handled using the L<attributes> pragma, or starting
4433from Perl 5.8.0 also via the L<Attribute::Handlers> module.  See
4434L<perlsub/"Private Variables via my()"> for details.
4435
4436=item next LABEL
4437X<next> X<continue>
4438
4439=item next EXPR
4440
4441=item next
4442
4443=for Pod::Functions iterate a block prematurely
4444
4445The L<C<next>|/next LABEL> command is like the C<continue> statement in
4446C; it starts the next iteration of the loop:
4447
4448    LINE: while (<STDIN>) {
4449        next LINE if /^#/;  # discard comments
4450        #...
4451    }
4452
4453Note that if there were a L<C<continue>|/continue BLOCK> block on the
4454above, it would get
4455executed even on discarded lines.  If LABEL is omitted, the command
4456refers to the innermost enclosing loop.  The C<next EXPR> form, available
4457as of Perl 5.18.0, allows a label name to be computed at run time, being
4458otherwise identical to C<next LABEL>.
4459
4460L<C<next>|/next LABEL> cannot return a value from a block that typically
4461returns a value, such as C<eval {}>, C<sub {}>, or C<do {}>. It will perform
4462its flow control behavior, which precludes any return value. It should not be
4463used to exit a L<C<grep>|/grep BLOCK LIST> or L<C<map>|/map BLOCK LIST>
4464operation.
4465
4466Note that a block by itself is semantically identical to a loop
4467that executes once.  Thus L<C<next>|/next LABEL> will exit such a block
4468early.
4469
4470See also L<C<continue>|/continue BLOCK> for an illustration of how
4471L<C<last>|/last LABEL>, L<C<next>|/next LABEL>, and
4472L<C<redo>|/redo LABEL> work.
4473
4474Unlike most named operators, this has the same precedence as assignment.
4475It is also exempt from the looks-like-a-function rule, so
4476C<next ("foo")."bar"> will cause "bar" to be part of the argument to
4477L<C<next>|/next LABEL>.
4478
4479=item no MODULE VERSION LIST
4480X<no declarations>
4481X<unimporting>
4482
4483=item no MODULE VERSION
4484
4485=item no MODULE LIST
4486
4487=item no MODULE
4488
4489=item no VERSION
4490
4491=for Pod::Functions unimport some module symbols or semantics at compile time
4492
4493See the L<C<use>|/use Module VERSION LIST> function, of which
4494L<C<no>|/no MODULE VERSION LIST> is the opposite.
4495
4496=item oct EXPR
4497X<oct> X<octal> X<hex> X<hexadecimal> X<binary> X<bin>
4498
4499=item oct
4500
4501=for Pod::Functions convert a string to an octal number
4502
4503Interprets EXPR as an octal string and returns the corresponding
4504value.  An octal string consists of octal digits and, as of Perl 5.33.5,
4505an optional C<0o> or C<o> prefix.  Each octal digit may be preceded by
4506a single underscore, which will be ignored.
4507(If EXPR happens to start off with C<0x> or C<x>, interprets it as a
4508hex string.  If EXPR starts off with C<0b> or C<b>, it is interpreted as a
4509binary string.  Leading whitespace is ignored in all three cases.)
4510The following will handle decimal, binary, octal, and hex in standard
4511Perl notation:
4512
4513    $val = oct($val) if $val =~ /^0/;
4514
4515If EXPR is omitted, uses L<C<$_>|perlvar/$_>.   To go the other way
4516(produce a number in octal), use L<C<sprintf>|/sprintf FORMAT, LIST> or
4517L<C<printf>|/printf FILEHANDLE FORMAT, LIST>:
4518
4519    my $dec_perms = (stat("filename"))[2] & 07777;
4520    my $oct_perm_str = sprintf "%o", $perms;
4521
4522The L<C<oct>|/oct EXPR> function is commonly used when a string such as
4523C<644> needs
4524to be converted into a file mode, for example.  Although Perl
4525automatically converts strings into numbers as needed, this automatic
4526conversion assumes base 10.
4527
4528Leading white space is ignored without warning, as too are any trailing
4529non-digits, such as a decimal point (L<C<oct>|/oct EXPR> only handles
4530non-negative integers, not negative integers or floating point).
4531
4532=item open FILEHANDLE,MODE,EXPR
4533X<open> X<pipe> X<file, open> X<fopen>
4534
4535=item open FILEHANDLE,MODE,EXPR,LIST
4536
4537=item open FILEHANDLE,MODE,REFERENCE
4538
4539=item open FILEHANDLE,EXPR
4540
4541=item open FILEHANDLE
4542
4543=for Pod::Functions open a file, pipe, or descriptor
4544
4545Associates an internal FILEHANDLE with the external file specified by
4546EXPR. That filehandle will subsequently allow you to perform
4547I/O operations on that file, such as reading from it or writing to it.
4548
4549Instead of a filename, you may specify an external command
4550(plus an optional argument list) or a scalar reference, in order to open
4551filehandles on commands or in-memory scalars, respectively.
4552
4553A thorough reference to C<open> follows. For a gentler introduction to
4554the basics of C<open>, see also the L<perlopentut> manual page.
4555
4556=over
4557
4558=item Working with files
4559
4560Most often, C<open> gets invoked with three arguments: the required
4561FILEHANDLE (usually an empty scalar variable), followed by MODE (usually
4562a literal describing the I/O mode the filehandle will use), and then the
4563filename  that the new filehandle will refer to.
4564
4565=over
4566
4567=item Simple examples
4568
4569Reading from a file:
4570
4571    open(my $fh, "<", "input.txt")
4572        or die "Can't open < input.txt: $!";
4573
4574    # Process every line in input.txt
4575    while (my $line = <$fh>) {
4576        #
4577        # ... do something interesting with $line here ...
4578        #
4579    }
4580
4581or writing to one:
4582
4583    open(my $fh, ">", "output.txt")
4584        or die "Can't open > output.txt: $!";
4585
4586    print $fh "This line gets printed into output.txt.\n";
4587
4588For a summary of common filehandle operations such as these, see
4589L<perlintro/Files and I/O>.
4590
4591=item About filehandles
4592
4593The first argument to C<open>, labeled FILEHANDLE in this reference, is
4594usually a scalar variable. (Exceptions exist, described in "Other
4595considerations", below.) If the call to C<open> succeeds, then the
4596expression provided as FILEHANDLE will get assigned an open
4597I<filehandle>. That filehandle provides an internal reference to the
4598specified external file, conveniently stored in a Perl variable, and
4599ready for I/O operations such as reading and writing.
4600
4601=item About modes
4602
4603When calling C<open> with three or more arguments, the second argument
4604-- labeled MODE here -- defines the I<open mode>. MODE is usually a
4605literal string comprising special characters that define the intended
4606I/O role of the filehandle being created: whether it's read-only, or
4607read-and-write, and so on.
4608
4609If MODE is C<< < >>, the file is opened for input (read-only).
4610If MODE is C<< > >>, the file is opened for output, with existing files
4611first being truncated ("clobbered") and nonexisting files newly created.
4612If MODE is C<<< >> >>>, the file is opened for appending, again being
4613created if necessary.
4614
4615You can put a C<+> in front of the C<< > >> or C<< < >> to
4616indicate that you want both read and write access to the file; thus
4617C<< +< >> is almost always preferred for read/write updates--the
4618C<< +> >> mode would clobber the file first.  You can't usually use
4619either read-write mode for updating textfiles, since they have
4620variable-length records.  See the B<-i> switch in
4621L<perlrun|perlrun/-i[extension]> for a better approach.  The file is
4622created with permissions of C<0666> modified by the process's
4623L<C<umask>|/umask EXPR> value.
4624
4625These various prefixes correspond to the L<fopen(3)> modes of C<r>,
4626C<r+>, C<w>, C<w+>, C<a>, and C<a+>.
4627
4628More examples of different modes in action:
4629
4630 # Open a file for concatenation
4631 open(my $log, ">>", "/usr/spool/news/twitlog")
4632     or warn "Couldn't open log file; discarding input";
4633
4634 # Open a file for reading and writing
4635 open(my $dbase, "+<", "dbase.mine")
4636     or die "Can't open 'dbase.mine' for update: $!";
4637
4638=item Checking the return value
4639
4640Open returns nonzero on success, the undefined value otherwise.  If the
4641C<open> involved a pipe, the return value happens to be the pid of the
4642subprocess.
4643
4644When opening a file, it's seldom a good idea to continue if the request
4645failed, so C<open> is frequently used with L<C<die>|/die LIST>. Even if
4646you want your code to do something other than C<die> on a failed open,
4647you should still always check the return value from opening a file.
4648
4649=back
4650
4651=item Specifying I/O layers in MODE
4652
4653You can use the three-argument form of open to specify
4654I/O layers (sometimes referred to as "disciplines") to apply to the new
4655filehandle. These affect how the input and output are processed (see
4656L<open> and
4657L<PerlIO> for more details).  For example:
4658
4659    # loads PerlIO::encoding automatically
4660    open(my $fh, "<:encoding(UTF-8)", $filename)
4661        || die "Can't open UTF-8 encoded $filename: $!";
4662
4663This opens the UTF8-encoded file containing Unicode characters;
4664see L<perluniintro>.  Note that if layers are specified in the
4665three-argument form, then default layers stored in
4666L<C<${^OPEN}>|perlvar/${^OPEN}>
4667(usually set by the L<open> pragma or the switch C<-CioD>) are ignored.
4668Those layers will also be ignored if you specify a colon with no name
4669following it.  In that case the default layer for the operating system
4670(:raw on Unix, :crlf on Windows) is used.
4671
4672On some systems (in general, DOS- and Windows-based systems)
4673L<C<binmode>|/binmode FILEHANDLE, LAYER> is necessary when you're not
4674working with a text file.  For the sake of portability it is a good idea
4675always to use it when appropriate, and never to use it when it isn't
4676appropriate.  Also, people can set their I/O to be by default
4677UTF8-encoded Unicode, not bytes.
4678
4679=item Using C<undef> for temporary files
4680
4681As a special case the three-argument form with a read/write mode and the third
4682argument being L<C<undef>|/undef EXPR>:
4683
4684    open(my $tmp, "+>", undef) or die ...
4685
4686opens a filehandle to a newly created empty anonymous temporary file.
4687(This happens under any mode, which makes C<< +> >> the only useful and
4688sensible mode to use.)  You will need to
4689L<C<seek>|/seek FILEHANDLE,POSITION,WHENCE> to do the reading.
4690
4691
4692=item Opening a filehandle into an in-memory scalar
4693
4694You can open filehandles directly to Perl scalars instead of a file or
4695other resource external to the program. To do so, provide a reference to
4696that scalar as the third argument to C<open>, like so:
4697
4698 open(my $memory, ">", \$var)
4699     or die "Can't open memory file: $!";
4700 print $memory "foo!\n";    # output will appear in $var
4701
4702To (re)open C<STDOUT> or C<STDERR> as an in-memory file, close it first:
4703
4704    close STDOUT;
4705    open(STDOUT, ">", \$variable)
4706	or die "Can't open STDOUT: $!";
4707
4708The scalars for in-memory files are treated as octet strings: unless
4709the file is being opened with truncation the scalar may not contain
4710any code points over 0xFF.
4711
4712Opening in-memory files I<can> fail for a variety of reasons.  As with
4713any other C<open>, check the return value for success.
4714
4715I<Technical note>: This feature works only when Perl is built with
4716PerlIO -- the default, except with older (pre-5.16) Perl installations
4717that were configured to not include it (e.g. via C<Configure
4718-Uuseperlio>). You can see whether your Perl was built with PerlIO by
4719running C<perl -V:useperlio>.  If it says C<'define'>, you have PerlIO;
4720otherwise you don't.
4721
4722See L<perliol> for detailed info on PerlIO.
4723
4724=item Opening a filehandle into a command
4725
4726If MODE is C<|->, then the filename is
4727interpreted as a command to which output is to be piped, and if MODE
4728is C<-|>, the filename is interpreted as a command that pipes
4729output to us.  In the two-argument (and one-argument) form, one should
4730replace dash (C<->) with the command.
4731See L<perlipc/"Using open() for IPC"> for more examples of this.
4732(You are not allowed to L<C<open>|/open FILEHANDLE,MODE,EXPR> to a command
4733that pipes both in I<and> out, but see L<IPC::Open2>, L<IPC::Open3>, and
4734L<perlipc/"Bidirectional Communication with Another Process"> for
4735alternatives.)
4736
4737
4738 open(my $article_fh, "-|", "caesar <$article")  # decrypt
4739                                                 # article
4740     or die "Can't start caesar: $!";
4741
4742 open(my $article_fh, "caesar <$article |")      # ditto
4743     or die "Can't start caesar: $!";
4744
4745 open(my $out_fh, "|-", "sort >Tmp$$")    # $$ is our process id
4746     or die "Can't start sort: $!";
4747
4748
4749In the form of pipe opens taking three or more arguments, if LIST is specified
4750(extra arguments after the command name) then LIST becomes arguments
4751to the command invoked if the platform supports it.  The meaning of
4752L<C<open>|/open FILEHANDLE,MODE,EXPR> with more than three arguments for
4753non-pipe modes is not yet defined, but experimental "layers" may give
4754extra LIST arguments meaning.
4755
4756If you open a pipe on the command C<-> (that is, specify either C<|-> or C<-|>
4757with the one- or two-argument forms of
4758L<C<open>|/open FILEHANDLE,MODE,EXPR>), an implicit L<C<fork>|/fork> is done,
4759so L<C<open>|/open FILEHANDLE,MODE,EXPR> returns twice: in the parent process
4760it returns the pid
4761of the child process, and in the child process it returns (a defined) C<0>.
4762Use C<defined($pid)> or C<//> to determine whether the open was successful.
4763
4764For example, use either
4765
4766   my $child_pid = open(my $from_kid, "-|")
4767        // die "Can't fork: $!";
4768
4769or
4770
4771   my $child_pid = open(my $to_kid,   "|-")
4772        // die "Can't fork: $!";
4773
4774followed by
4775
4776    if ($child_pid) {
4777	# am the parent:
4778	# either write $to_kid or else read $from_kid
4779	...
4780       waitpid $child_pid, 0;
4781    } else {
4782	# am the child; use STDIN/STDOUT normally
4783	...
4784	exit;
4785    }
4786
4787The filehandle behaves normally for the parent, but I/O to that
4788filehandle is piped from/to the STDOUT/STDIN of the child process.
4789In the child process, the filehandle isn't opened--I/O happens from/to
4790the new STDOUT/STDIN.  Typically this is used like the normal
4791piped open when you want to exercise more control over just how the
4792pipe command gets executed, such as when running setuid and
4793you don't want to have to scan shell commands for metacharacters.
4794
4795The following blocks are more or less equivalent:
4796
4797    open(my $fh, "|tr '[a-z]' '[A-Z]'");
4798    open(my $fh, "|-", "tr '[a-z]' '[A-Z]'");
4799    open(my $fh, "|-") || exec 'tr', '[a-z]', '[A-Z]';
4800    open(my $fh, "|-", "tr", '[a-z]', '[A-Z]');
4801
4802    open(my $fh, "cat -n '$file'|");
4803    open(my $fh, "-|", "cat -n '$file'");
4804    open(my $fh, "-|") || exec "cat", "-n", $file;
4805    open(my $fh, "-|", "cat", "-n", $file);
4806
4807The last two examples in each block show the pipe as "list form", which
4808is not yet supported on all platforms. (If your platform has a real
4809L<C<fork>|/fork>, such as Linux and macOS, you can use the list form; it
4810also works on Windows with Perl 5.22 or later.) You would want to use
4811the list form of the pipe so you can pass literal arguments to the
4812command without risk of the shell interpreting any shell metacharacters
4813in them. However, this also bars you from opening pipes to commands that
4814intentionally contain shell metacharacters, such as:
4815
4816    open(my $fh, "|cat -n | expand -4 | lpr")
4817    	|| die "Can't open pipeline to lpr: $!";
4818
4819See L<perlipc/"Safe Pipe Opens"> for more examples of this.
4820
4821=item Duping filehandles
4822
4823You may also, in the Bourne shell tradition, specify an EXPR beginning
4824with C<< >& >>, in which case the rest of the string is interpreted
4825as the name of a filehandle (or file descriptor, if numeric) to be
4826duped (as in L<dup(2)>) and opened.  You may use C<&> after C<< > >>,
4827C<<< >> >>>, C<< < >>, C<< +> >>, C<<< +>> >>>, and C<< +< >>.
4828The mode you specify should match the mode of the original filehandle.
4829(Duping a filehandle does not take into account any existing contents
4830of IO buffers.)  If you use the three-argument
4831form, then you can pass either a
4832number, the name of a filehandle, or the normal "reference to a glob".
4833
4834Here is a script that saves, redirects, and restores C<STDOUT> and
4835C<STDERR> using various methods:
4836
4837    #!/usr/bin/perl
4838    open(my $oldout, ">&STDOUT")
4839        or die "Can't dup STDOUT: $!";
4840    open(OLDERR,     ">&", \*STDERR)
4841        or die "Can't dup STDERR: $!";
4842
4843    open(STDOUT, '>', "foo.out")
4844        or die "Can't redirect STDOUT: $!";
4845    open(STDERR, ">&STDOUT")
4846        or die "Can't dup STDOUT: $!";
4847
4848    select STDERR; $| = 1;  # make unbuffered
4849    select STDOUT; $| = 1;  # make unbuffered
4850
4851    print STDOUT "stdout 1\n";  # this works for
4852    print STDERR "stderr 1\n";  # subprocesses too
4853
4854    open(STDOUT, ">&", $oldout)
4855        or die "Can't dup \$oldout: $!";
4856    open(STDERR, ">&OLDERR")
4857        or die "Can't dup OLDERR: $!";
4858
4859    print STDOUT "stdout 2\n";
4860    print STDERR "stderr 2\n";
4861
4862If you specify C<< '<&=X' >>, where C<X> is a file descriptor number
4863or a filehandle, then Perl will do an equivalent of C's L<fdopen(3)> of
4864that file descriptor (and not call L<dup(2)>); this is more
4865parsimonious of file descriptors.  For example:
4866
4867    # open for input, reusing the fileno of $fd
4868    open(my $fh, "<&=", $fd)
4869
4870or
4871
4872    open(my $fh, "<&=$fd")
4873
4874or
4875
4876    # open for append, using the fileno of $oldfh
4877    open(my $fh, ">>&=", $oldfh)
4878
4879Being parsimonious on filehandles is also useful (besides being
4880parsimonious) for example when something is dependent on file
4881descriptors, like for example locking using
4882L<C<flock>|/flock FILEHANDLE,OPERATION>.  If you do just
4883C<< open(my $A, ">>&", $B) >>, the filehandle C<$A> will not have the
4884same file descriptor as C<$B>, and therefore C<flock($A)> will not
4885C<flock($B)> nor vice versa.  But with C<< open(my $A, ">>&=", $B) >>,
4886the filehandles will share the same underlying system file descriptor.
4887
4888Note that under Perls older than 5.8.0, Perl uses the standard C library's'
4889L<fdopen(3)> to implement the C<=> functionality.  On many Unix systems,
4890L<fdopen(3)> fails when file descriptors exceed a certain value, typically 255.
4891For Perls 5.8.0 and later, PerlIO is (most often) the default.
4892
4893=item Legacy usage
4894
4895This section describes ways to call C<open> outside of best practices;
4896you may encounter these uses in older code. Perl does not consider their
4897use deprecated, exactly, but neither is it recommended in new code, for
4898the sake of clarity and readability.
4899
4900=over
4901
4902=item Specifying mode and filename as a single argument
4903
4904In the one- and two-argument forms of the call, the mode and filename
4905should be concatenated (in that order), preferably separated by white
4906space.  You can--but shouldn't--omit the mode in these forms when that mode
4907is C<< < >>.  It is safe to use the two-argument form of
4908L<C<open>|/open FILEHANDLE,MODE,EXPR> if the filename argument is a known literal.
4909
4910 open(my $dbase, "+<dbase.mine")          # ditto
4911     or die "Can't open 'dbase.mine' for update: $!";
4912
4913In the two-argument (and one-argument) form, opening C<< <- >>
4914or C<-> opens STDIN and opening C<< >- >> opens STDOUT.
4915
4916New code should favor the three-argument form of C<open> over this older
4917form. Declaring the mode and the filename as two distinct arguments
4918avoids any confusion between the two.
4919
4920=item Calling C<open> with one argument via global variables
4921
4922As a shortcut, a one-argument call takes the filename from the global
4923scalar variable of the same name as the filehandle:
4924
4925    $ARTICLE = 100;
4926    open(ARTICLE)
4927        or die "Can't find article $ARTICLE: $!\n";
4928
4929Here C<$ARTICLE> must be a global (package) scalar variable - not one
4930declared with L<C<my>|/my VARLIST> or L<C<state>|/state VARLIST>.
4931
4932=item Assigning a filehandle to a bareword
4933
4934An older style is to use a bareword as the filehandle, as
4935
4936    open(FH, "<", "input.txt")
4937       or die "Can't open < input.txt: $!";
4938
4939Then you can use C<FH> as the filehandle, in C<< close FH >> and C<<
4940<FH> >> and so on.  Note that it's a global variable, so this form is
4941not recommended when dealing with filehandles other than Perl's built-in ones
4942(e.g. STDOUT and STDIN).  In fact, using a bareword for the filehandle is
4943an error when the C<bareword_filehandles> feature has been disabled.  This
4944feature is disabled by default when in the scope of C<use v5.36.0> or later.
4945
4946
4947=back
4948
4949=item Other considerations
4950
4951=over
4952
4953=item Automatic filehandle closure
4954
4955The filehandle will be closed when its reference count reaches zero. If
4956it is a lexically scoped variable declared with L<C<my>|/my VARLIST>,
4957that usually means the end of the enclosing scope.  However, this
4958automatic close does not check for errors, so it is better to explicitly
4959close filehandles, especially those used for writing:
4960
4961    close($handle)
4962       || warn "close failed: $!";
4963
4964=item Automatic pipe flushing
4965
4966Perl will attempt to flush all files opened for
4967output before any operation that may do a fork, but this may not be
4968supported on some platforms (see L<perlport>).  To be safe, you may need
4969to set L<C<$E<verbar>>|perlvar/$E<verbar>> (C<$AUTOFLUSH> in L<English>)
4970or call the C<autoflush> method of L<C<IO::Handle>|IO::Handle/METHODS>
4971on any open handles.
4972
4973On systems that support a close-on-exec flag on files, the flag will
4974be set for the newly opened file descriptor as determined by the value
4975of L<C<$^F>|perlvar/$^F>.  See L<perlvar/$^F>.
4976
4977Closing any piped filehandle causes the parent process to wait for the
4978child to finish, then returns the status value in L<C<$?>|perlvar/$?> and
4979L<C<${^CHILD_ERROR_NATIVE}>|perlvar/${^CHILD_ERROR_NATIVE}>.
4980
4981=item Direct versus by-reference assignment of filehandles
4982
4983If FILEHANDLE -- the first argument in a call to C<open> -- is an
4984undefined scalar variable (or array or hash element), a new filehandle
4985is autovivified, meaning that the variable is assigned a reference to a
4986newly allocated anonymous filehandle.  Otherwise if FILEHANDLE is an
4987expression, its value is the real filehandle.  (This is considered a
4988symbolic reference, so C<use strict "refs"> should I<not> be in effect.)
4989
4990=item Whitespace and special characters in the filename argument
4991
4992The filename passed to the one- and two-argument forms of
4993L<C<open>|/open FILEHANDLE,MODE,EXPR> will
4994have leading and trailing whitespace deleted and normal
4995redirection characters honored.  This property, known as "magic open",
4996can often be used to good effect.  A user could specify a filename of
4997F<"rsh cat file |">, or you could change certain filenames as needed:
4998
4999    $filename =~ s/(.*\.gz)\s*$/gzip -dc < $1|/;
5000    open(my $fh, $filename)
5001        or die "Can't open $filename: $!";
5002
5003Use the three-argument form to open a file with arbitrary weird characters in it,
5004
5005    open(my $fh, "<", $file)
5006    	|| die "Can't open $file: $!";
5007
5008otherwise it's necessary to protect any leading and trailing whitespace:
5009
5010    $file =~ s#^(\s)#./$1#;
5011    open(my $fh, "< $file\0")
5012    	|| die "Can't open $file: $!";
5013
5014(this may not work on some bizarre filesystems).  One should
5015conscientiously choose between the I<magic> and I<three-argument> form
5016of L<C<open>|/open FILEHANDLE,MODE,EXPR>:
5017
5018    open(my $in, $ARGV[0]) || die "Can't open $ARGV[0]: $!";
5019
5020will allow the user to specify an argument of the form C<"rsh cat file |">,
5021but will not work on a filename that happens to have a trailing space, while
5022
5023    open(my $in, "<", $ARGV[0])
5024    	|| die "Can't open $ARGV[0]: $!";
5025
5026will have exactly the opposite restrictions. (However, some shells
5027support the syntax C<< perl your_program.pl <( rsh cat file ) >>, which
5028produces a filename that can be opened normally.)
5029
5030=item Invoking C-style C<open>
5031
5032If you want a "real" C L<open(2)>, then you should use the
5033L<C<sysopen>|/sysopen FILEHANDLE,FILENAME,MODE> function, which involves
5034no such magic (but uses different filemodes than Perl
5035L<C<open>|/open FILEHANDLE,MODE,EXPR>, which corresponds to C L<fopen(3)>).
5036This is another way to protect your filenames from interpretation.  For
5037example:
5038
5039    use IO::Handle;
5040    sysopen(my $fh, $path, O_RDWR|O_CREAT|O_EXCL)
5041        or die "Can't open $path: $!";
5042    $fh->autoflush(1);
5043    print $fh "stuff $$\n";
5044    seek($fh, 0, 0);
5045    print "File contains: ", readline($fh);
5046
5047See L<C<seek>|/seek FILEHANDLE,POSITION,WHENCE> for some details about
5048mixing reading and writing.
5049
5050=item Portability issues
5051
5052See L<perlport/open>.
5053
5054=back
5055
5056=back
5057
5058
5059=item opendir DIRHANDLE,EXPR
5060X<opendir>
5061
5062=for Pod::Functions open a directory
5063
5064Opens a directory named EXPR for processing by
5065L<C<readdir>|/readdir DIRHANDLE>, L<C<telldir>|/telldir DIRHANDLE>,
5066L<C<seekdir>|/seekdir DIRHANDLE,POS>,
5067L<C<rewinddir>|/rewinddir DIRHANDLE>, and
5068L<C<closedir>|/closedir DIRHANDLE>.  Returns true if successful.
5069DIRHANDLE may be an expression whose value can be used as an indirect
5070dirhandle, usually the real dirhandle name.  If DIRHANDLE is an undefined
5071scalar variable (or array or hash element), the variable is assigned a
5072reference to a new anonymous dirhandle; that is, it's autovivified.
5073Dirhandles are the same objects as filehandles; an I/O object can only
5074be open as one of these handle types at once.
5075
5076See the example at L<C<readdir>|/readdir DIRHANDLE>.
5077
5078=item ord EXPR
5079X<ord> X<encoding>
5080
5081=item ord
5082
5083=for Pod::Functions find a character's numeric representation
5084
5085Returns the numeric value of the first character of EXPR.
5086If EXPR is an empty string, returns 0.  If EXPR is omitted, uses
5087L<C<$_>|perlvar/$_>.
5088(Note I<character>, not byte.)
5089
5090For the reverse, see L<C<chr>|/chr NUMBER>.
5091See L<perlunicode> for more about Unicode.
5092
5093=item our VARLIST
5094X<our> X<global>
5095
5096=item our TYPE VARLIST
5097
5098=item our VARLIST : ATTRS
5099
5100=item our TYPE VARLIST : ATTRS
5101
5102=for Pod::Functions +5.6.0 declare and assign a package variable (lexical scoping)
5103
5104L<C<our>|/our VARLIST> makes a lexical alias to a package (i.e. global)
5105variable of the same name in the current package for use within the
5106current lexical scope.
5107
5108L<C<our>|/our VARLIST> has the same scoping rules as
5109L<C<my>|/my VARLIST> or L<C<state>|/state VARLIST>, meaning that it is
5110only valid within a lexical scope.  Unlike L<C<my>|/my VARLIST> and
5111L<C<state>|/state VARLIST>, which both declare new (lexical) variables,
5112L<C<our>|/our VARLIST> only creates an alias to an existing variable: a
5113package variable of the same name.
5114
5115This means that when C<use strict 'vars'> is in effect, L<C<our>|/our
5116VARLIST> lets you use a package variable without qualifying it with the
5117package name, but only within the lexical scope of the
5118L<C<our>|/our VARLIST> declaration.  This applies immediately--even
5119within the same statement.
5120
5121    package Foo;
5122    use v5.36;  # which implies "use strict;"
5123
5124    $Foo::foo = 23;
5125
5126    {
5127        our $foo;   # alias to $Foo::foo
5128        print $foo; # prints 23
5129    }
5130
5131    print $Foo::foo; # prints 23
5132
5133    print $foo; # ERROR: requires explicit package name
5134
5135This works even if the package variable has not been used before, as
5136package variables spring into existence when first used.
5137
5138    package Foo;
5139    use v5.36;
5140
5141    our $foo = 23;   # just like $Foo::foo = 23
5142
5143    print $Foo::foo; # prints 23
5144
5145Because the variable becomes legal immediately under C<use strict 'vars'>, so
5146long as there is no variable with that name is already in scope, you can then
5147reference the package variable again even within the same statement.
5148
5149    package Foo;
5150    use v5.36;
5151
5152    my  $foo = $foo; # error, undeclared $foo on right-hand side
5153    our $foo = $foo; # no errors
5154
5155If more than one variable is listed, the list must be placed
5156in parentheses.
5157
5158    our($bar, $baz);
5159
5160An L<C<our>|/our VARLIST> declaration declares an alias for a package
5161variable that will be visible
5162across its entire lexical scope, even across package boundaries.  The
5163package in which the variable is entered is determined at the point
5164of the declaration, not at the point of use.  This means the following
5165behavior holds:
5166
5167    package Foo;
5168    our $bar;      # declares $Foo::bar for rest of lexical scope
5169    $bar = 20;
5170
5171    package Bar;
5172    print $bar;    # prints 20, as it refers to $Foo::bar
5173
5174Multiple L<C<our>|/our VARLIST> declarations with the same name in the
5175same lexical
5176scope are allowed if they are in different packages.  If they happen
5177to be in the same package, Perl will emit warnings if you have asked
5178for them, just like multiple L<C<my>|/my VARLIST> declarations.  Unlike
5179a second L<C<my>|/my VARLIST> declaration, which will bind the name to a
5180fresh variable, a second L<C<our>|/our VARLIST> declaration in the same
5181package, in the same scope, is merely redundant.
5182
5183    use warnings;
5184    package Foo;
5185    our $bar;      # declares $Foo::bar for rest of lexical scope
5186    $bar = 20;
5187
5188    package Bar;
5189    our $bar = 30; # declares $Bar::bar for rest of lexical scope
5190    print $bar;    # prints 30
5191
5192    our $bar;      # emits warning but has no other effect
5193    print $bar;    # still prints 30
5194
5195An L<C<our>|/our VARLIST> declaration may also have a list of attributes
5196associated with it.
5197
5198The exact semantics and interface of TYPE and ATTRS are still
5199evolving.  TYPE is currently bound to the use of the L<fields> pragma,
5200and attributes are handled using the L<attributes> pragma, or, starting
5201from Perl 5.8.0, also via the L<Attribute::Handlers> module.  See
5202L<perlsub/"Private Variables via my()"> for details.
5203
5204Note that with a parenthesised list, L<C<undef>|/undef EXPR> can be used
5205as a dummy placeholder, for example to skip assignment of initial
5206values:
5207
5208    our ( undef, $min, $hour ) = localtime;
5209
5210L<C<our>|/our VARLIST> differs from L<C<use vars>|vars>, which allows
5211use of an unqualified name I<only> within the affected package, but
5212across scopes.
5213
5214=item pack TEMPLATE,LIST
5215X<pack>
5216
5217=for Pod::Functions convert a list into a binary representation
5218
5219Takes a LIST of values and converts it into a string using the rules
5220given by the TEMPLATE.  The resulting string is the concatenation of
5221the converted values.  Typically, each converted value looks
5222like its machine-level representation.  For example, on 32-bit machines
5223an integer may be represented by a sequence of 4 bytes, which  will in
5224Perl be presented as a string that's 4 characters long.
5225
5226See L<perlpacktut> for an introduction to this function.
5227
5228The TEMPLATE is a sequence of characters that give the order and type
5229of values, as follows:
5230
5231    a  A string with arbitrary binary data, will be null padded.
5232    A  A text (ASCII) string, will be space padded.
5233    Z  A null-terminated (ASCIZ) string, will be null padded.
5234
5235    b  A bit string (ascending bit order inside each byte,
5236       like vec()).
5237    B  A bit string (descending bit order inside each byte).
5238    h  A hex string (low nybble first).
5239    H  A hex string (high nybble first).
5240
5241    c  A signed char (8-bit) value.
5242    C  An unsigned char (octet) value.
5243    W  An unsigned char value (can be greater than 255).
5244
5245    s  A signed short (16-bit) value.
5246    S  An unsigned short value.
5247
5248    l  A signed long (32-bit) value.
5249    L  An unsigned long value.
5250
5251    q  A signed quad (64-bit) value.
5252    Q  An unsigned quad value.
5253         (Quads are available only if your system supports 64-bit
5254          integer values _and_ if Perl has been compiled to support
5255          those.  Raises an exception otherwise.)
5256
5257    i  A signed integer value.
5258    I  An unsigned integer value.
5259         (This 'integer' is _at_least_ 32 bits wide.  Its exact
5260          size depends on what a local C compiler calls 'int'.)
5261
5262    n  An unsigned short (16-bit) in "network" (big-endian) order.
5263    N  An unsigned long (32-bit) in "network" (big-endian) order.
5264    v  An unsigned short (16-bit) in "VAX" (little-endian) order.
5265    V  An unsigned long (32-bit) in "VAX" (little-endian) order.
5266
5267    j  A Perl internal signed integer value (IV).
5268    J  A Perl internal unsigned integer value (UV).
5269
5270    f  A single-precision float in native format.
5271    d  A double-precision float in native format.
5272
5273    F  A Perl internal floating-point value (NV) in native format
5274    D  A float of long-double precision in native format.
5275         (Long doubles are available only if your system supports
5276          long double values. Raises an exception otherwise.
5277          Note that there are different long double formats.)
5278
5279    p  A pointer to a null-terminated string.
5280    P  A pointer to a structure (fixed-length string).
5281
5282    u  A uuencoded string.
5283    U  A Unicode character number.  Encodes to a character in char-
5284       acter mode and UTF-8 (or UTF-EBCDIC in EBCDIC platforms) in
5285       byte mode.  Also on EBCDIC platforms, the character number will
5286       be the native EBCDIC value for character numbers below 256.
5287       This allows most programs using this feature to not have to
5288       care which type of platform they are running on.
5289
5290    w  A BER compressed integer (not an ASN.1 BER, see perlpacktut
5291       for details).  Its bytes represent an unsigned integer in
5292       base 128, most significant digit first, with as few digits
5293       as possible.  Bit eight (the high bit) is set on each byte
5294       except the last.
5295
5296    x  A null byte (a.k.a ASCII NUL, "\000", chr(0))
5297    X  Back up a byte.
5298    @  Null-fill or truncate to absolute position, counted from the
5299       start of the innermost ()-group.
5300    .  Null-fill or truncate to absolute position specified by
5301       the value.
5302    (  Start of a ()-group.
5303
5304One or more modifiers below may optionally follow certain letters in the
5305TEMPLATE (the second column lists letters for which the modifier is valid):
5306
5307    !   sSlLiI     Forces native (short, long, int) sizes instead
5308                   of fixed (16-/32-bit) sizes.
5309
5310    !   xX         Make x and X act as alignment commands.
5311
5312    !   nNvV       Treat integers as signed instead of unsigned.
5313
5314    !   @.         Specify position as byte offset in the internal
5315                   representation of the packed string.  Efficient
5316                   but dangerous.
5317
5318    >   sSiIlLqQ   Force big-endian byte-order on the type.
5319        jJfFdDpP   (The "big end" touches the construct.)
5320
5321    <   sSiIlLqQ   Force little-endian byte-order on the type.
5322        jJfFdDpP   (The "little end" touches the construct.)
5323
5324The C<< > >> and C<< < >> modifiers can also be used on C<()> groups
5325to force a particular byte-order on all components in that group,
5326including all its subgroups.
5327
5328=begin comment
5329
5330Larry recalls that the hex and bit string formats (H, h, B, b) were added to
5331pack for processing data from NASA's Magellan probe.  Magellan was in an
5332elliptical orbit, using the antenna for the radar mapping when close to
5333Venus and for communicating data back to Earth for the rest of the orbit.
5334There were two transmission units, but one of these failed, and then the
5335other developed a fault whereby it would randomly flip the sense of all the
5336bits. It was easy to automatically detect complete records with the correct
5337sense, and complete records with all the bits flipped. However, this didn't
5338recover the records where the sense flipped midway. A colleague of Larry's
5339was able to pretty much eyeball where the records flipped, so they wrote an
5340editor named kybble (a pun on the dog food Kibbles 'n Bits) to enable him to
5341manually correct the records and recover the data. For this purpose pack
5342gained the hex and bit string format specifiers.
5343
5344git shows that they were added to perl 3.0 in patch #44 (Jan 1991, commit
534527e2fb84680b9cc1), but the patch description makes no mention of their
5346addition, let alone the story behind them.
5347
5348=end comment
5349
5350The following rules apply:
5351
5352=over
5353
5354=item *
5355
5356Each letter may optionally be followed by a number indicating the repeat
5357count.  A numeric repeat count may optionally be enclosed in brackets, as
5358in C<pack("C[80]", @arr)>.  The repeat count gobbles that many values from
5359the LIST when used with all format types other than C<a>, C<A>, C<Z>, C<b>,
5360C<B>, C<h>, C<H>, C<@>, C<.>, C<x>, C<X>, and C<P>, where it means
5361something else, described below.  Supplying a C<*> for the repeat count
5362instead of a number means to use however many items are left, except for:
5363
5364=over
5365
5366=item *
5367
5368C<@>, C<x>, and C<X>, where it is equivalent to C<0>.
5369
5370=item *
5371
5372<.>, where it means relative to the start of the string.
5373
5374=item *
5375
5376C<u>, where it is equivalent to 1 (or 45, which here is equivalent).
5377
5378=back
5379
5380One can replace a numeric repeat count with a template letter enclosed in
5381brackets to use the packed byte length of the bracketed template for the
5382repeat count.
5383
5384For example, the template C<x[L]> skips as many bytes as in a packed long,
5385and the template C<"$t X[$t] $t"> unpacks twice whatever $t (when
5386variable-expanded) unpacks.  If the template in brackets contains alignment
5387commands (such as C<x![d]>), its packed length is calculated as if the
5388start of the template had the maximal possible alignment.
5389
5390When used with C<Z>, a C<*> as the repeat count is guaranteed to add a
5391trailing null byte, so the resulting string is always one byte longer than
5392the byte length of the item itself.
5393
5394When used with C<@>, the repeat count represents an offset from the start
5395of the innermost C<()> group.
5396
5397When used with C<.>, the repeat count determines the starting position to
5398calculate the value offset as follows:
5399
5400=over
5401
5402=item *
5403
5404If the repeat count is C<0>, it's relative to the current position.
5405
5406=item *
5407
5408If the repeat count is C<*>, the offset is relative to the start of the
5409packed string.
5410
5411=item *
5412
5413And if it's an integer I<n>, the offset is relative to the start of the
5414I<n>th innermost C<( )> group, or to the start of the string if I<n> is
5415bigger then the group level.
5416
5417=back
5418
5419The repeat count for C<u> is interpreted as the maximal number of bytes
5420to encode per line of output, with 0, 1 and 2 replaced by 45.  The repeat
5421count should not be more than 65.
5422
5423=item *
5424
5425The C<a>, C<A>, and C<Z> types gobble just one value, but pack it as a
5426string of length count, padding with nulls or spaces as needed.  When
5427unpacking, C<A> strips trailing whitespace and nulls, C<Z> strips everything
5428after the first null, and C<a> returns data with no stripping at all.
5429
5430If the value to pack is too long, the result is truncated.  If it's too
5431long and an explicit count is provided, C<Z> packs only C<$count-1> bytes,
5432followed by a null byte.  Thus C<Z> always packs a trailing null, except
5433when the count is 0.
5434
5435=item *
5436
5437Likewise, the C<b> and C<B> formats pack a string that's that many bits long.
5438Each such format generates 1 bit of the result.  These are typically followed
5439by a repeat count like C<B8> or C<B64>.
5440
5441Each result bit is based on the least-significant bit of the corresponding
5442input character, i.e., on C<ord($char)%2>.  In particular, characters C<"0">
5443and C<"1"> generate bits 0 and 1, as do characters C<"\000"> and C<"\001">.
5444
5445Starting from the beginning of the input string, each 8-tuple
5446of characters is converted to 1 character of output.  With format C<b>,
5447the first character of the 8-tuple determines the least-significant bit of a
5448character; with format C<B>, it determines the most-significant bit of
5449a character.
5450
5451If the length of the input string is not evenly divisible by 8, the
5452remainder is packed as if the input string were padded by null characters
5453at the end.  Similarly during unpacking, "extra" bits are ignored.
5454
5455If the input string is longer than needed, remaining characters are ignored.
5456
5457A C<*> for the repeat count uses all characters of the input field.
5458On unpacking, bits are converted to a string of C<0>s and C<1>s.
5459
5460=item *
5461
5462The C<h> and C<H> formats pack a string that many nybbles (4-bit groups,
5463representable as hexadecimal digits, C<"0".."9"> C<"a".."f">) long.
5464
5465For each such format, L<C<pack>|/pack TEMPLATE,LIST> generates 4 bits of result.
5466With non-alphabetical characters, the result is based on the 4 least-significant
5467bits of the input character, i.e., on C<ord($char)%16>.  In particular,
5468characters C<"0"> and C<"1"> generate nybbles 0 and 1, as do bytes
5469C<"\000"> and C<"\001">.  For characters C<"a".."f"> and C<"A".."F">, the result
5470is compatible with the usual hexadecimal digits, so that C<"a"> and
5471C<"A"> both generate the nybble C<0xA==10>.  Use only these specific hex
5472characters with this format.
5473
5474Starting from the beginning of the template to
5475L<C<pack>|/pack TEMPLATE,LIST>, each pair
5476of characters is converted to 1 character of output.  With format C<h>, the
5477first character of the pair determines the least-significant nybble of the
5478output character; with format C<H>, it determines the most-significant
5479nybble.
5480
5481If the length of the input string is not even, it behaves as if padded by
5482a null character at the end.  Similarly, "extra" nybbles are ignored during
5483unpacking.
5484
5485If the input string is longer than needed, extra characters are ignored.
5486
5487A C<*> for the repeat count uses all characters of the input field.  For
5488L<C<unpack>|/unpack TEMPLATE,EXPR>, nybbles are converted to a string of
5489hexadecimal digits.
5490
5491=item *
5492
5493The C<p> format packs a pointer to a null-terminated string.  You are
5494responsible for ensuring that the string is not a temporary value, as that
5495could potentially get deallocated before you got around to using the packed
5496result.  The C<P> format packs a pointer to a structure of the size indicated
5497by the length.  A null pointer is created if the corresponding value for
5498C<p> or C<P> is L<C<undef>|/undef EXPR>; similarly with
5499L<C<unpack>|/unpack TEMPLATE,EXPR>, where a null pointer unpacks into
5500L<C<undef>|/undef EXPR>.
5501
5502If your system has a strange pointer size--meaning a pointer is neither as
5503big as an int nor as big as a long--it may not be possible to pack or
5504unpack pointers in big- or little-endian byte order.  Attempting to do
5505so raises an exception.
5506
5507=item *
5508
5509The C</> template character allows packing and unpacking of a sequence of
5510items where the packed structure contains a packed item count followed by
5511the packed items themselves.  This is useful when the structure you're
5512unpacking has encoded the sizes or repeat counts for some of its fields
5513within the structure itself as separate fields.
5514
5515For L<C<pack>|/pack TEMPLATE,LIST>, you write
5516I<length-item>C</>I<sequence-item>, and the
5517I<length-item> describes how the length value is packed.  Formats likely
5518to be of most use are integer-packing ones like C<n> for Java strings,
5519C<w> for ASN.1 or SNMP, and C<N> for Sun XDR.
5520
5521For L<C<pack>|/pack TEMPLATE,LIST>, I<sequence-item> may have a repeat
5522count, in which case
5523the minimum of that and the number of available items is used as the argument
5524for I<length-item>.  If it has no repeat count or uses a '*', the number
5525of available items is used.
5526
5527For L<C<unpack>|/unpack TEMPLATE,EXPR>, an internal stack of integer
5528arguments unpacked so far is
5529used.  You write C</>I<sequence-item> and the repeat count is obtained by
5530popping off the last element from the stack.  The I<sequence-item> must not
5531have a repeat count.
5532
5533If I<sequence-item> refers to a string type (C<"A">, C<"a">, or C<"Z">),
5534the I<length-item> is the string length, not the number of strings.  With
5535an explicit repeat count for pack, the packed string is adjusted to that
5536length.  For example:
5537
5538 This code:                             gives this result:
5539
5540 unpack("W/a", "\004Gurusamy")          ("Guru")
5541 unpack("a3/A A*", "007 Bond  J ")      (" Bond", "J")
5542 unpack("a3 x2 /A A*", "007: Bond, J.") ("Bond, J", ".")
5543
5544 pack("n/a* w/a","hello,","world")     "\000\006hello,\005world"
5545 pack("a/W2", ord("a") .. ord("z"))    "2ab"
5546
5547The I<length-item> is not returned explicitly from
5548L<C<unpack>|/unpack TEMPLATE,EXPR>.
5549
5550Supplying a count to the I<length-item> format letter is only useful with
5551C<A>, C<a>, or C<Z>.  Packing with a I<length-item> of C<a> or C<Z> may
5552introduce C<"\000"> characters, which Perl does not regard as legal in
5553numeric strings.
5554
5555=item *
5556
5557The integer types C<s>, C<S>, C<l>, and C<L> may be
5558followed by a C<!> modifier to specify native shorts or
5559longs.  As shown in the example above, a bare C<l> means
5560exactly 32 bits, although the native C<long> as seen by the local C compiler
5561may be larger.  This is mainly an issue on 64-bit platforms.  You can
5562see whether using C<!> makes any difference this way:
5563
5564    printf "format s is %d, s! is %d\n",
5565	length pack("s"), length pack("s!");
5566
5567    printf "format l is %d, l! is %d\n",
5568	length pack("l"), length pack("l!");
5569
5570
5571C<i!> and C<I!> are also allowed, but only for completeness' sake:
5572they are identical to C<i> and C<I>.
5573
5574The actual sizes (in bytes) of native shorts, ints, longs, and long
5575longs on the platform where Perl was built are also available from
5576the command line:
5577
5578    $ perl -V:{short,int,long{,long}}size
5579    shortsize='2';
5580    intsize='4';
5581    longsize='4';
5582    longlongsize='8';
5583
5584or programmatically via the L<C<Config>|Config> module:
5585
5586       use Config;
5587       print $Config{shortsize},    "\n";
5588       print $Config{intsize},      "\n";
5589       print $Config{longsize},     "\n";
5590       print $Config{longlongsize}, "\n";
5591
5592C<$Config{longlongsize}> is undefined on systems without
5593long long support.
5594
5595=item *
5596
5597The integer formats C<s>, C<S>, C<i>, C<I>, C<l>, C<L>, C<j>, and C<J> are
5598inherently non-portable between processors and operating systems because
5599they obey native byteorder and endianness.  For example, a 4-byte integer
56000x12345678 (305419896 decimal) would be ordered natively (arranged in and
5601handled by the CPU registers) into bytes as
5602
5603    0x12 0x34 0x56 0x78  # big-endian
5604    0x78 0x56 0x34 0x12  # little-endian
5605
5606Basically, Intel and VAX CPUs are little-endian, while everybody else,
5607including Motorola m68k/88k, PPC, Sparc, HP PA, Power, and Cray, are
5608big-endian.  Alpha and MIPS can be either: Digital/Compaq uses (well, used)
5609them in little-endian mode, but SGI/Cray uses them in big-endian mode.
5610
5611The names I<big-endian> and I<little-endian> are comic references to the
5612egg-eating habits of the little-endian Lilliputians and the big-endian
5613Blefuscudians from the classic Jonathan Swift satire, I<Gulliver's Travels>.
5614This entered computer lingo via the paper "On Holy Wars and a Plea for
5615Peace" by Danny Cohen, USC/ISI IEN 137, April 1, 1980.
5616
5617Some systems may have even weirder byte orders such as
5618
5619   0x56 0x78 0x12 0x34
5620   0x34 0x12 0x78 0x56
5621
5622These are called mid-endian, middle-endian, mixed-endian, or just weird.
5623
5624You can determine your system endianness with this incantation:
5625
5626   printf("%#02x ", $_) for unpack("W*", pack L=>0x12345678);
5627
5628The byteorder on the platform where Perl was built is also available
5629via L<Config>:
5630
5631    use Config;
5632    print "$Config{byteorder}\n";
5633
5634or from the command line:
5635
5636    $ perl -V:byteorder
5637
5638Byteorders C<"1234"> and C<"12345678"> are little-endian; C<"4321">
5639and C<"87654321"> are big-endian.  Systems with multiarchitecture binaries
5640will have C<"ffff">, signifying that static information doesn't work,
5641one must use runtime probing.
5642
5643For portably packed integers, either use the formats C<n>, C<N>, C<v>,
5644and C<V> or else use the C<< > >> and C<< < >> modifiers described
5645immediately below.  See also L<perlport>.
5646
5647=item *
5648
5649Also floating point numbers have endianness.  Usually (but not always)
5650this agrees with the integer endianness.  Even though most platforms
5651these days use the IEEE 754 binary format, there are differences,
5652especially if the long doubles are involved.  You can see the
5653C<Config> variables C<doublekind> and C<longdblkind> (also C<doublesize>,
5654C<longdblsize>): the "kind" values are enums, unlike C<byteorder>.
5655
5656Portability-wise the best option is probably to keep to the IEEE 754
565764-bit doubles, and of agreed-upon endianness.  Another possibility
5658is the C<"%a">) format of L<C<printf>|/printf FILEHANDLE FORMAT, LIST>.
5659
5660=item *
5661
5662Starting with Perl 5.10.0, integer and floating-point formats, along with
5663the C<p> and C<P> formats and C<()> groups, may all be followed by the
5664C<< > >> or C<< < >> endianness modifiers to respectively enforce big-
5665or little-endian byte-order.  These modifiers are especially useful
5666given how C<n>, C<N>, C<v>, and C<V> don't cover signed integers,
566764-bit integers, or floating-point values.
5668
5669Here are some concerns to keep in mind when using an endianness modifier:
5670
5671=over
5672
5673=item *
5674
5675Exchanging signed integers between different platforms works only
5676when all platforms store them in the same format.  Most platforms store
5677signed integers in two's-complement notation, so usually this is not an issue.
5678
5679=item *
5680
5681The C<< > >> or C<< < >> modifiers can only be used on floating-point
5682formats on big- or little-endian machines.  Otherwise, attempting to
5683use them raises an exception.
5684
5685=item *
5686
5687Forcing big- or little-endian byte-order on floating-point values for
5688data exchange can work only if all platforms use the same
5689binary representation such as IEEE floating-point.  Even if all
5690platforms are using IEEE, there may still be subtle differences.  Being able
5691to use C<< > >> or C<< < >> on floating-point values can be useful,
5692but also dangerous if you don't know exactly what you're doing.
5693It is not a general way to portably store floating-point values.
5694
5695=item *
5696
5697When using C<< > >> or C<< < >> on a C<()> group, this affects
5698all types inside the group that accept byte-order modifiers,
5699including all subgroups.  It is silently ignored for all other
5700types.  You are not allowed to override the byte-order within a group
5701that already has a byte-order modifier suffix.
5702
5703=back
5704
5705=item *
5706
5707Real numbers (floats and doubles) are in native machine format only.
5708Due to the multiplicity of floating-point formats and the lack of a
5709standard "network" representation for them, no facility for interchange has been
5710made.  This means that packed floating-point data written on one machine
5711may not be readable on another, even if both use IEEE floating-point
5712arithmetic (because the endianness of the memory representation is not part
5713of the IEEE spec).  See also L<perlport>.
5714
5715If you know I<exactly> what you're doing, you can use the C<< > >> or C<< < >>
5716modifiers to force big- or little-endian byte-order on floating-point values.
5717
5718Because Perl uses doubles (or long doubles, if configured) internally for
5719all numeric calculation, converting from double into float and thence
5720to double again loses precision, so C<unpack("f", pack("f", $foo)>)
5721will not in general equal $foo.
5722
5723=item *
5724
5725Pack and unpack can operate in two modes: character mode (C<C0> mode) where
5726the packed string is processed per character, and UTF-8 byte mode (C<U0> mode)
5727where the packed string is processed in its UTF-8-encoded Unicode form on
5728a byte-by-byte basis.  Character mode is the default
5729unless the format string starts with C<U>.  You
5730can always switch mode mid-format with an explicit
5731C<C0> or C<U0> in the format.  This mode remains in effect until the next
5732mode change, or until the end of the C<()> group it (directly) applies to.
5733
5734Using C<C0> to get Unicode characters while using C<U0> to get I<non>-Unicode
5735bytes is not necessarily obvious.   Probably only the first of these
5736is what you want:
5737
5738    $ perl -CS -E 'say "\x{3B1}\x{3C9}"' |
5739      perl -CS -ne 'printf "%v04X\n", $_ for unpack("C0A*", $_)'
5740    03B1.03C9
5741    $ perl -CS -E 'say "\x{3B1}\x{3C9}"' |
5742      perl -CS -ne 'printf "%v02X\n", $_ for unpack("U0A*", $_)'
5743    CE.B1.CF.89
5744    $ perl -CS -E 'say "\x{3B1}\x{3C9}"' |
5745      perl -C0 -ne 'printf "%v02X\n", $_ for unpack("C0A*", $_)'
5746    CE.B1.CF.89
5747    $ perl -CS -E 'say "\x{3B1}\x{3C9}"' |
5748      perl -C0 -ne 'printf "%v02X\n", $_ for unpack("U0A*", $_)'
5749    C3.8E.C2.B1.C3.8F.C2.89
5750
5751Those examples also illustrate that you should not try to use
5752L<C<pack>|/pack TEMPLATE,LIST>/L<C<unpack>|/unpack TEMPLATE,EXPR> as a
5753substitute for the L<Encode> module.
5754
5755=item *
5756
5757You must yourself do any alignment or padding by inserting, for example,
5758enough C<"x">es while packing.  There is no way for
5759L<C<pack>|/pack TEMPLATE,LIST> and L<C<unpack>|/unpack TEMPLATE,EXPR>
5760to know where characters are going to or coming from, so they
5761handle their output and input as flat sequences of characters.
5762
5763=item *
5764
5765A C<()> group is a sub-TEMPLATE enclosed in parentheses.  A group may
5766take a repeat count either as postfix, or for
5767L<C<unpack>|/unpack TEMPLATE,EXPR>, also via the C</>
5768template character.  Within each repetition of a group, positioning with
5769C<@> starts over at 0.  Therefore, the result of
5770
5771    pack("@1A((@2A)@3A)", qw[X Y Z])
5772
5773is the string C<"\0X\0\0YZ">.
5774
5775=item *
5776
5777C<x> and C<X> accept the C<!> modifier to act as alignment commands: they
5778jump forward or back to the closest position aligned at a multiple of C<count>
5779characters.  For example, to L<C<pack>|/pack TEMPLATE,LIST> or
5780L<C<unpack>|/unpack TEMPLATE,EXPR> a C structure like
5781
5782    struct {
5783	char   c;    /* one signed, 8-bit character */
5784	double d;
5785	char   cc[2];
5786    }
5787
5788one may need to use the template C<c x![d] d c[2]>.  This assumes that
5789doubles must be aligned to the size of double.
5790
5791For alignment commands, a C<count> of 0 is equivalent to a C<count> of 1;
5792both are no-ops.
5793
5794=item *
5795
5796C<n>, C<N>, C<v> and C<V> accept the C<!> modifier to
5797represent signed 16-/32-bit integers in big-/little-endian order.
5798This is portable only when all platforms sharing packed data use the
5799same binary representation for signed integers; for example, when all
5800platforms use two's-complement representation.
5801
5802=item *
5803
5804Comments can be embedded in a TEMPLATE using C<#> through the end of line.
5805White space can separate pack codes from each other, but modifiers and
5806repeat counts must follow immediately.  Breaking complex templates into
5807individual line-by-line components, suitably annotated, can do as much to
5808improve legibility and maintainability of pack/unpack formats as C</x> can
5809for complicated pattern matches.
5810
5811=item *
5812
5813If TEMPLATE requires more arguments than L<C<pack>|/pack TEMPLATE,LIST>
5814is given, L<C<pack>|/pack TEMPLATE,LIST>
5815assumes additional C<""> arguments.  If TEMPLATE requires fewer arguments
5816than given, extra arguments are ignored.
5817
5818=item *
5819
5820Attempting to pack the special floating point values C<Inf> and C<NaN>
5821(infinity, also in negative, and not-a-number) into packed integer values
5822(like C<"L">) is a fatal error.  The reason for this is that there simply
5823isn't any sensible mapping for these special values into integers.
5824
5825=back
5826
5827Examples:
5828
5829    $foo = pack("WWWW",65,66,67,68);
5830    # foo eq "ABCD"
5831    $foo = pack("W4",65,66,67,68);
5832    # same thing
5833    $foo = pack("W4",0x24b6,0x24b7,0x24b8,0x24b9);
5834    # same thing with Unicode circled letters.
5835    $foo = pack("U4",0x24b6,0x24b7,0x24b8,0x24b9);
5836    # same thing with Unicode circled letters.  You don't get the
5837    # UTF-8 bytes because the U at the start of the format caused
5838    # a switch to U0-mode, so the UTF-8 bytes get joined into
5839    # characters
5840    $foo = pack("C0U4",0x24b6,0x24b7,0x24b8,0x24b9);
5841    # foo eq "\xe2\x92\xb6\xe2\x92\xb7\xe2\x92\xb8\xe2\x92\xb9"
5842    # This is the UTF-8 encoding of the string in the
5843    # previous example
5844
5845    $foo = pack("ccxxcc",65,66,67,68);
5846    # foo eq "AB\0\0CD"
5847
5848    # NOTE: The examples above featuring "W" and "c" are true
5849    # only on ASCII and ASCII-derived systems such as ISO Latin 1
5850    # and UTF-8.  On EBCDIC systems, the first example would be
5851    #      $foo = pack("WWWW",193,194,195,196);
5852
5853    $foo = pack("s2",1,2);
5854    # "\001\000\002\000" on little-endian
5855    # "\000\001\000\002" on big-endian
5856
5857    $foo = pack("a4","abcd","x","y","z");
5858    # "abcd"
5859
5860    $foo = pack("aaaa","abcd","x","y","z");
5861    # "axyz"
5862
5863    $foo = pack("a14","abcdefg");
5864    # "abcdefg\0\0\0\0\0\0\0"
5865
5866    $foo = pack("i9pl", gmtime);
5867    # a real struct tm (on my system anyway)
5868
5869    $utmp_template = "Z8 Z8 Z16 L";
5870    $utmp = pack($utmp_template, @utmp1);
5871    # a struct utmp (BSDish)
5872
5873    @utmp2 = unpack($utmp_template, $utmp);
5874    # "@utmp1" eq "@utmp2"
5875
5876    sub bintodec {
5877        unpack("N", pack("B32", substr("0" x 32 . shift, -32)));
5878    }
5879
5880    $foo = pack('sx2l', 12, 34);
5881    # short 12, two zero bytes padding, long 34
5882    $bar = pack('s@4l', 12, 34);
5883    # short 12, zero fill to position 4, long 34
5884    # $foo eq $bar
5885    $baz = pack('s.l', 12, 4, 34);
5886    # short 12, zero fill to position 4, long 34
5887
5888    $foo = pack('nN', 42, 4711);
5889    # pack big-endian 16- and 32-bit unsigned integers
5890    $foo = pack('S>L>', 42, 4711);
5891    # exactly the same
5892    $foo = pack('s<l<', -42, 4711);
5893    # pack little-endian 16- and 32-bit signed integers
5894    $foo = pack('(sl)<', -42, 4711);
5895    # exactly the same
5896
5897The same template may generally also be used in
5898L<C<unpack>|/unpack TEMPLATE,EXPR>.
5899
5900=item package NAMESPACE
5901
5902=item package NAMESPACE VERSION
5903X<package> X<module> X<namespace> X<version>
5904
5905=item package NAMESPACE BLOCK
5906
5907=item package NAMESPACE VERSION BLOCK
5908X<package> X<module> X<namespace> X<version>
5909
5910=for Pod::Functions declare a separate global namespace
5911
5912Declares the BLOCK or the rest of the compilation unit as being in the
5913given namespace.  The scope of the package declaration is either the
5914supplied code BLOCK or, in the absence of a BLOCK, from the declaration
5915itself through the end of current scope (the enclosing block, file, or
5916L<C<eval>|/eval EXPR>).  That is, the forms without a BLOCK are
5917operative through the end of the current scope, just like the
5918L<C<my>|/my VARLIST>, L<C<state>|/state VARLIST>, and
5919L<C<our>|/our VARLIST> operators.  All unqualified dynamic identifiers
5920in this scope will be in the given namespace, except where overridden by
5921another L<C<package>|/package NAMESPACE> declaration or
5922when they're one of the special identifiers that qualify into C<main::>,
5923like C<STDOUT>, C<ARGV>, C<ENV>, and the punctuation variables.
5924
5925A package statement affects dynamic variables only, including those
5926you've used L<C<local>|/local EXPR> on, but I<not> lexically-scoped
5927variables, which are created with L<C<my>|/my VARLIST>,
5928L<C<state>|/state VARLIST>, or L<C<our>|/our VARLIST>.  Typically it
5929would be the first declaration in a file included by
5930L<C<require>|/require VERSION> or L<C<use>|/use Module VERSION LIST>.
5931You can switch into a
5932package in more than one place, since this only determines which default
5933symbol table the compiler uses for the rest of that block.  You can refer to
5934identifiers in other packages than the current one by prefixing the identifier
5935with the package name and a double colon, as in C<$SomePack::var>
5936or C<ThatPack::INPUT_HANDLE>.  If package name is omitted, the C<main>
5937package is assumed.  That is, C<$::sail> is equivalent to
5938C<$main::sail> (as well as to C<$main'sail>, still seen in ancient
5939code, mostly from Perl 4).
5940
5941If VERSION is provided, L<C<package>|/package NAMESPACE> sets the
5942C<$VERSION> variable in the given
5943namespace to a L<version> object with the VERSION provided.  VERSION must be a
5944"strict" style version number as defined by the L<version> module: a positive
5945decimal number (integer or decimal-fraction) without exponentiation or else a
5946dotted-decimal v-string with a leading 'v' character and at least three
5947components.  You should set C<$VERSION> only once per package.
5948
5949See L<perlmod/"Packages"> for more information about packages, modules,
5950and classes.  See L<perlsub> for other scoping issues.
5951
5952=item __PACKAGE__
5953X<__PACKAGE__>
5954
5955=for Pod::Functions +5.004 the current package
5956
5957A special token that returns the name of the package in which it occurs.
5958
5959=item pipe READHANDLE,WRITEHANDLE
5960X<pipe>
5961
5962=for Pod::Functions open a pair of connected filehandles
5963
5964Opens a pair of connected pipes like the corresponding system call.
5965Note that if you set up a loop of piped processes, deadlock can occur
5966unless you are very careful.  In addition, note that Perl's pipes use
5967IO buffering, so you may need to set L<C<$E<verbar>>|perlvar/$E<verbar>>
5968to flush your WRITEHANDLE after each command, depending on the
5969application.
5970
5971Returns true on success.
5972
5973See L<IPC::Open2>, L<IPC::Open3>, and
5974L<perlipc/"Bidirectional Communication with Another Process">
5975for examples of such things.
5976
5977On systems that support a close-on-exec flag on files, that flag is set
5978on all newly opened file descriptors whose
5979L<C<fileno>|/fileno FILEHANDLE>s are I<higher> than the current value of
5980L<C<$^F>|perlvar/$^F> (by default 2 for C<STDERR>).  See L<perlvar/$^F>.
5981
5982=item pop ARRAY
5983X<pop> X<stack>
5984
5985=item pop
5986
5987=for Pod::Functions remove the last element from an array and return it
5988
5989Pops and returns the last value of the array, shortening the array by
5990one element.
5991
5992Returns the undefined value if the array is empty, although this may
5993also happen at other times.  If ARRAY is omitted, pops the
5994L<C<@ARGV>|perlvar/@ARGV> array in the main program, but the
5995L<C<@_>|perlvar/@_> array in subroutines, just like
5996L<C<shift>|/shift ARRAY>.
5997
5998Starting with Perl 5.14, an experimental feature allowed
5999L<C<pop>|/pop ARRAY> to take a
6000scalar expression. This experiment has been deemed unsuccessful, and was
6001removed as of Perl 5.24.
6002
6003=item pos SCALAR
6004X<pos> X<match, position>
6005
6006=item pos
6007
6008=for Pod::Functions find or set the offset for the last/next m//g search
6009
6010Returns the offset of where the last C<m//g> search left off for the
6011variable in question (L<C<$_>|perlvar/$_> is used when the variable is not
6012specified).  This offset is in characters unless the
6013(no-longer-recommended) L<C<use bytes>|bytes> pragma is in effect, in
6014which case the offset is in bytes.  Note that 0 is a valid match offset.
6015L<C<undef>|/undef EXPR> indicates
6016that the search position is reset (usually due to match failure, but
6017can also be because no match has yet been run on the scalar).
6018
6019L<C<pos>|/pos SCALAR> directly accesses the location used by the regexp
6020engine to store the offset, so assigning to L<C<pos>|/pos SCALAR> will
6021change that offset, and so will also influence the C<\G> zero-width
6022assertion in regular expressions.  Both of these effects take place for
6023the next match, so you can't affect the position with
6024L<C<pos>|/pos SCALAR> during the current match, such as in
6025C<(?{pos() = 5})> or C<s//pos() = 5/e>.
6026
6027Setting L<C<pos>|/pos SCALAR> also resets the I<matched with
6028zero-length> flag, described
6029under L<perlre/"Repeated Patterns Matching a Zero-length Substring">.
6030
6031Because a failed C<m//gc> match doesn't reset the offset, the return
6032from L<C<pos>|/pos SCALAR> won't change either in this case.  See
6033L<perlre> and L<perlop>.
6034
6035=item print FILEHANDLE LIST
6036X<print>
6037
6038=item print FILEHANDLE
6039
6040=item print LIST
6041
6042=item print
6043
6044=for Pod::Functions output a list to a filehandle
6045
6046Prints a string or a list of strings.  Returns true if successful.
6047FILEHANDLE may be a scalar variable containing the name of or a reference
6048to the filehandle, thus introducing one level of indirection.  (NOTE: If
6049FILEHANDLE is a variable and the next token is a term, it may be
6050misinterpreted as an operator unless you interpose a C<+> or put
6051parentheses around the arguments.)  If FILEHANDLE is omitted, prints to the
6052last selected (see L<C<select>|/select FILEHANDLE>) output handle.  If
6053LIST is omitted, prints L<C<$_>|perlvar/$_> to the currently selected
6054output handle.  To use FILEHANDLE alone to print the content of
6055L<C<$_>|perlvar/$_> to it, you must use a bareword filehandle like
6056C<FH>, not an indirect one like C<$fh>.  To set the default output handle
6057to something other than STDOUT, use the select operation.
6058
6059The current value of L<C<$,>|perlvar/$,> (if any) is printed between
6060each LIST item.  The current value of L<C<$\>|perlvar/$\> (if any) is
6061printed after the entire LIST has been printed.  Because print takes a
6062LIST, anything in the LIST is evaluated in list context, including any
6063subroutines whose return lists you pass to
6064L<C<print>|/print FILEHANDLE LIST>.  Be careful not to follow the print
6065keyword with a left
6066parenthesis unless you want the corresponding right parenthesis to
6067terminate the arguments to the print; put parentheses around all arguments
6068(or interpose a C<+>, but that doesn't look as good).
6069
6070If you're storing handles in an array or hash, or in general whenever
6071you're using any expression more complex than a bareword handle or a plain,
6072unsubscripted scalar variable to retrieve it, you will have to use a block
6073returning the filehandle value instead, in which case the LIST may not be
6074omitted:
6075
6076    print { $files[$i] } "stuff\n";
6077    print { $OK ? *STDOUT : *STDERR } "stuff\n";
6078
6079Printing to a closed pipe or socket will generate a SIGPIPE signal.  See
6080L<perlipc> for more on signal handling.
6081
6082=item printf FILEHANDLE FORMAT, LIST
6083X<printf>
6084
6085=item printf FILEHANDLE
6086
6087=item printf FORMAT, LIST
6088
6089=item printf
6090
6091=for Pod::Functions output a formatted list to a filehandle
6092
6093Equivalent to C<print FILEHANDLE sprintf(FORMAT, LIST)>, except that
6094L<C<$\>|perlvar/$\> (the output record separator) is not appended.  The
6095FORMAT and the LIST are actually parsed as a single list.  The first
6096argument of the list will be interpreted as the
6097L<C<printf>|/printf FILEHANDLE FORMAT, LIST> format.  This means that
6098C<printf(@_)> will use C<$_[0]> as the format.  See
6099L<sprintf|/sprintf FORMAT, LIST> for an explanation of the format
6100argument.  If C<use locale> (including C<use locale ':not_characters'>)
6101is in effect and L<C<POSIX::setlocale>|POSIX/C<setlocale>> has been
6102called, the character used for the decimal separator in formatted
6103floating-point numbers is affected by the C<LC_NUMERIC> locale setting.
6104See L<perllocale> and L<POSIX>.
6105
6106For historical reasons, if you omit the list, L<C<$_>|perlvar/$_> is
6107used as the format;
6108to use FILEHANDLE without a list, you must use a bareword filehandle like
6109C<FH>, not an indirect one like C<$fh>.  However, this will rarely do what
6110you want; if L<C<$_>|perlvar/$_> contains formatting codes, they will be
6111replaced with the empty string and a warning will be emitted if
6112L<warnings> are enabled.  Just use L<C<print>|/print FILEHANDLE LIST> if
6113you want to print the contents of L<C<$_>|perlvar/$_>.
6114
6115Don't fall into the trap of using a
6116L<C<printf>|/printf FILEHANDLE FORMAT, LIST> when a simple
6117L<C<print>|/print FILEHANDLE LIST> would do.  The
6118L<C<print>|/print FILEHANDLE LIST> is more efficient and less error
6119prone.
6120
6121=item prototype FUNCTION
6122X<prototype>
6123
6124=item prototype
6125
6126=for Pod::Functions +5.002 get the prototype (if any) of a subroutine
6127
6128Returns the prototype of a function as a string (or
6129L<C<undef>|/undef EXPR> if the
6130function has no prototype).  FUNCTION is a reference to, or the name of,
6131the function whose prototype you want to retrieve.  If FUNCTION is omitted,
6132L<C<$_>|perlvar/$_> is used.
6133
6134If FUNCTION is a string starting with C<CORE::>, the rest is taken as a
6135name for a Perl builtin.  If the builtin's arguments
6136cannot be adequately expressed by a prototype
6137(such as L<C<system>|/system LIST>), L<C<prototype>|/prototype FUNCTION>
6138returns L<C<undef>|/undef EXPR>, because the builtin
6139does not really behave like a Perl function.  Otherwise, the string
6140describing the equivalent prototype is returned.
6141
6142=item push ARRAY,LIST
6143X<push> X<stack>
6144
6145=for Pod::Functions append one or more elements to an array
6146
6147Treats ARRAY as a stack by appending the values of LIST to the end of
6148ARRAY.  The length of ARRAY increases by the length of LIST.  Has the same
6149effect as
6150
6151    for my $value (LIST) {
6152        $ARRAY[++$#ARRAY] = $value;
6153    }
6154
6155but is more efficient.  Returns the number of elements in the array following
6156the completed L<C<push>|/push ARRAY,LIST>.
6157
6158Starting with Perl 5.14, an experimental feature allowed
6159L<C<push>|/push ARRAY,LIST> to take a
6160scalar expression. This experiment has been deemed unsuccessful, and was
6161removed as of Perl 5.24.
6162
6163=item q/STRING/
6164
6165=for Pod::Functions singly quote a string
6166
6167=item qq/STRING/
6168
6169=for Pod::Functions doubly quote a string
6170
6171=item qw/STRING/
6172
6173=for Pod::Functions quote a list of words
6174
6175=item qx/STRING/
6176
6177=for Pod::Functions backquote quote a string
6178
6179Generalized quotes.  See L<perlop/"Quote-Like Operators">.
6180
6181=item qr/STRING/
6182
6183=for Pod::Functions +5.005 compile pattern
6184
6185Regexp-like quote.  See L<perlop/"Regexp Quote-Like Operators">.
6186
6187=item quotemeta EXPR
6188X<quotemeta> X<metacharacter>
6189
6190=item quotemeta
6191
6192=for Pod::Functions quote regular expression magic characters
6193
6194Returns the value of EXPR with all the ASCII non-"word"
6195characters backslashed.  (That is, all ASCII characters not matching
6196C</[A-Za-z_0-9]/> will be preceded by a backslash in the
6197returned string, regardless of any locale settings.)
6198This is the internal function implementing
6199the C<\Q> escape in double-quoted strings.
6200(See below for the behavior on non-ASCII code points.)
6201
6202If EXPR is omitted, uses L<C<$_>|perlvar/$_>.
6203
6204quotemeta (and C<\Q> ... C<\E>) are useful when interpolating strings into
6205regular expressions, because by default an interpolated variable will be
6206considered a mini-regular expression.  For example:
6207
6208    my $sentence = 'The quick brown fox jumped over the lazy dog';
6209    my $substring = 'quick.*?fox';
6210    $sentence =~ s{$substring}{big bad wolf};
6211
6212Will cause C<$sentence> to become C<'The big bad wolf jumped over...'>.
6213
6214On the other hand:
6215
6216    my $sentence = 'The quick brown fox jumped over the lazy dog';
6217    my $substring = 'quick.*?fox';
6218    $sentence =~ s{\Q$substring\E}{big bad wolf};
6219
6220Or:
6221
6222    my $sentence = 'The quick brown fox jumped over the lazy dog';
6223    my $substring = 'quick.*?fox';
6224    my $quoted_substring = quotemeta($substring);
6225    $sentence =~ s{$quoted_substring}{big bad wolf};
6226
6227Will both leave the sentence as is.
6228Normally, when accepting literal string input from the user,
6229L<C<quotemeta>|/quotemeta EXPR> or C<\Q> must be used.
6230
6231Beware that if you put literal backslashes (those not inside
6232interpolated variables) between C<\Q> and C<\E>, double-quotish
6233backslash interpolation may lead to confusing results.  If you
6234I<need> to use literal backslashes within C<\Q...\E>,
6235consult L<perlop/"Gory details of parsing quoted constructs">.
6236
6237Because the result of S<C<"\Q I<STRING> \E">> has all metacharacters
6238quoted, there is no way to insert a literal C<$> or C<@> inside a
6239C<\Q\E> pair.  If protected by C<\>, C<$> will be quoted to become
6240C<"\\\$">; if not, it is interpreted as the start of an interpolated
6241scalar.
6242
6243In Perl v5.14, all non-ASCII characters are quoted in non-UTF-8-encoded
6244strings, but not quoted in UTF-8 strings.
6245
6246Starting in Perl v5.16, Perl adopted a Unicode-defined strategy for
6247quoting non-ASCII characters; the quoting of ASCII characters is
6248unchanged.
6249
6250Also unchanged is the quoting of non-UTF-8 strings when outside the
6251scope of a
6252L<C<use feature 'unicode_strings'>|feature/The 'unicode_strings' feature>,
6253which is to quote all
6254characters in the upper Latin1 range.  This provides complete backwards
6255compatibility for old programs which do not use Unicode.  (Note that
6256C<unicode_strings> is automatically enabled within the scope of a
6257S<C<use v5.12>> or greater.)
6258
6259Within the scope of L<C<use locale>|locale>, all non-ASCII Latin1 code
6260points
6261are quoted whether the string is encoded as UTF-8 or not.  As mentioned
6262above, locale does not affect the quoting of ASCII-range characters.
6263This protects against those locales where characters such as C<"|"> are
6264considered to be word characters.
6265
6266Otherwise, Perl quotes non-ASCII characters using an adaptation from
6267Unicode (see L<https://www.unicode.org/reports/tr31/>).
6268The only code points that are quoted are those that have any of the
6269Unicode properties:  Pattern_Syntax, Pattern_White_Space, White_Space,
6270Default_Ignorable_Code_Point, or General_Category=Control.
6271
6272Of these properties, the two important ones are Pattern_Syntax and
6273Pattern_White_Space.  They have been set up by Unicode for exactly this
6274purpose of deciding which characters in a regular expression pattern
6275should be quoted.  No character that can be in an identifier has these
6276properties.
6277
6278Perl promises, that if we ever add regular expression pattern
6279metacharacters to the dozen already defined
6280(C<\ E<verbar> ( ) [ { ^ $ * + ? .>), that we will only use ones that have the
6281Pattern_Syntax property.  Perl also promises, that if we ever add
6282characters that are considered to be white space in regular expressions
6283(currently mostly affected by C</x>), they will all have the
6284Pattern_White_Space property.
6285
6286Unicode promises that the set of code points that have these two
6287properties will never change, so something that is not quoted in v5.16
6288will never need to be quoted in any future Perl release.  (Not all the
6289code points that match Pattern_Syntax have actually had characters
6290assigned to them; so there is room to grow, but they are quoted
6291whether assigned or not.  Perl, of course, would never use an
6292unassigned code point as an actual metacharacter.)
6293
6294Quoting characters that have the other 3 properties is done to enhance
6295the readability of the regular expression and not because they actually
6296need to be quoted for regular expression purposes (characters with the
6297White_Space property are likely to be indistinguishable on the page or
6298screen from those with the Pattern_White_Space property; and the other
6299two properties contain non-printing characters).
6300
6301=item rand EXPR
6302X<rand> X<random>
6303
6304=item rand
6305
6306=for Pod::Functions retrieve the next pseudorandom number
6307
6308Returns a random fractional number greater than or equal to C<0> and less
6309than the value of EXPR.  (EXPR should be positive.)  If EXPR is
6310omitted, the value C<1> is used.  Currently EXPR with the value C<0> is
6311also special-cased as C<1> (this was undocumented before Perl 5.8.0
6312and is subject to change in future versions of Perl).  Automatically calls
6313L<C<srand>|/srand EXPR> unless L<C<srand>|/srand EXPR> has already been
6314called.  See also L<C<srand>|/srand EXPR>.
6315
6316Apply L<C<int>|/int EXPR> to the value returned by L<C<rand>|/rand EXPR>
6317if you want random integers instead of random fractional numbers.  For
6318example,
6319
6320    int(rand(10))
6321
6322returns a random integer between C<0> and C<9>, inclusive.
6323
6324(Note: If your rand function consistently returns numbers that are too
6325large or too small, then your version of Perl was probably compiled
6326with the wrong number of RANDBITS.)
6327
6328B<L<C<rand>|/rand EXPR> is not cryptographically secure.  You should not rely
6329on it in security-sensitive situations.>  As of this writing, a
6330number of third-party CPAN modules offer random number generators
6331intended by their authors to be cryptographically secure,
6332including: L<Data::Entropy>, L<Crypt::Random>, L<Math::Random::Secure>,
6333and L<Math::TrulyRandom>.
6334
6335=item read FILEHANDLE,SCALAR,LENGTH,OFFSET
6336X<read> X<file, read>
6337
6338=item read FILEHANDLE,SCALAR,LENGTH
6339
6340=for Pod::Functions fixed-length buffered input from a filehandle
6341
6342Attempts to read LENGTH I<characters> of data into variable SCALAR
6343from the specified FILEHANDLE.  Returns the number of characters
6344actually read, C<0> at end of file, or undef if there was an error (in
6345the latter case L<C<$!>|perlvar/$!> is also set).  SCALAR will be grown
6346or shrunk
6347so that the last character actually read is the last character of the
6348scalar after the read.
6349
6350An OFFSET may be specified to place the read data at some place in the
6351string other than the beginning.  A negative OFFSET specifies
6352placement at that many characters counting backwards from the end of
6353the string.  A positive OFFSET greater than the length of SCALAR
6354results in the string being padded to the required size with C<"\0">
6355bytes before the result of the read is appended.
6356
6357The call is implemented in terms of either Perl's or your system's native
6358L<fread(3)> library function, via the L<PerlIO> layers applied to the
6359handle.  To get a true L<read(2)> system call, see
6360L<sysread|/sysread FILEHANDLE,SCALAR,LENGTH,OFFSET>.
6361
6362Note the I<characters>: depending on the status of the filehandle,
6363either (8-bit) bytes or characters are read.  By default, all
6364filehandles operate on bytes, but for example if the filehandle has
6365been opened with the C<:utf8> I/O layer (see
6366L<C<open>|/open FILEHANDLE,MODE,EXPR>, and the L<open>
6367pragma), the I/O will operate on UTF8-encoded Unicode
6368characters, not bytes.  Similarly for the C<:encoding> layer:
6369in that case pretty much any characters can be read.
6370
6371=item readdir DIRHANDLE
6372X<readdir>
6373
6374=for Pod::Functions get a directory from a directory handle
6375
6376Returns the next directory entry for a directory opened by
6377L<C<opendir>|/opendir DIRHANDLE,EXPR>.
6378If used in list context, returns all the rest of the entries in the
6379directory.  If there are no more entries, returns the undefined value in
6380scalar context and the empty list in list context.
6381
6382If you're planning to filetest the return values out of a
6383L<C<readdir>|/readdir DIRHANDLE>, you'd better prepend the directory in
6384question.  Otherwise, because we didn't L<C<chdir>|/chdir EXPR> there,
6385it would have been testing the wrong file.
6386
6387    opendir(my $dh, $some_dir) || die "Can't opendir $some_dir: $!";
6388    my @dots = grep { /^\./ && -f "$some_dir/$_" } readdir($dh);
6389    closedir $dh;
6390
6391As of Perl 5.12 you can use a bare L<C<readdir>|/readdir DIRHANDLE> in a
6392C<while> loop, which will set L<C<$_>|perlvar/$_> on every iteration.
6393If either a C<readdir> expression or an explicit assignment of a
6394C<readdir> expression to a scalar is used as a C<while>/C<for> condition,
6395then the condition actually tests for definedness of the expression's
6396value, not for its regular truth value.
6397
6398    opendir(my $dh, $some_dir) || die "Can't open $some_dir: $!";
6399    while (readdir $dh) {
6400        print "$some_dir/$_\n";
6401    }
6402    closedir $dh;
6403
6404To avoid confusing would-be users of your code who are running earlier
6405versions of Perl with mysterious failures, put this sort of thing at the
6406top of your file to signal that your code will work I<only> on Perls of a
6407recent vintage:
6408
6409    use v5.12; # so readdir assigns to $_ in a lone while test
6410
6411=item readline EXPR
6412
6413=item readline
6414X<readline> X<gets> X<fgets>
6415
6416=for Pod::Functions fetch a record from a file
6417
6418Reads from the filehandle whose typeglob is contained in EXPR (or from
6419C<*ARGV> if EXPR is not provided).  In scalar context, each call reads and
6420returns the next line until end-of-file is reached, whereupon the
6421subsequent call returns L<C<undef>|/undef EXPR>.  In list context, reads
6422until end-of-file is reached and returns a list of lines.  Note that the
6423notion of "line" used here is whatever you may have defined with
6424L<C<$E<sol>>|perlvar/$E<sol>> (or C<$INPUT_RECORD_SEPARATOR> in
6425L<English>).  See L<perlvar/"$/">.
6426
6427When L<C<$E<sol>>|perlvar/$E<sol>> is set to L<C<undef>|/undef EXPR>,
6428when L<C<readline>|/readline EXPR> is in scalar context (i.e., file
6429slurp mode), and when an empty file is read, it returns C<''> the first
6430time, followed by L<C<undef>|/undef EXPR> subsequently.
6431
6432This is the internal function implementing the C<< <EXPR> >>
6433operator, but you can use it directly.  The C<< <EXPR> >>
6434operator is discussed in more detail in L<perlop/"I/O Operators">.
6435
6436    my $line = <STDIN>;
6437    my $line = readline(STDIN);    # same thing
6438
6439If L<C<readline>|/readline EXPR> encounters an operating system error,
6440L<C<$!>|perlvar/$!> will be set with the corresponding error message.
6441It can be helpful to check L<C<$!>|perlvar/$!> when you are reading from
6442filehandles you don't trust, such as a tty or a socket.  The following
6443example uses the operator form of L<C<readline>|/readline EXPR> and dies
6444if the result is not defined.
6445
6446    while ( ! eof($fh) ) {
6447        defined( $_ = readline $fh ) or die "readline failed: $!";
6448        ...
6449    }
6450
6451Note that you can't handle L<C<readline>|/readline EXPR> errors
6452that way with the C<ARGV> filehandle.  In that case, you have to open
6453each element of L<C<@ARGV>|perlvar/@ARGV> yourself since
6454L<C<eof>|/eof FILEHANDLE> handles C<ARGV> differently.
6455
6456    foreach my $arg (@ARGV) {
6457        open(my $fh, $arg) or warn "Can't open $arg: $!";
6458
6459        while ( ! eof($fh) ) {
6460            defined( $_ = readline $fh )
6461                or die "readline failed for $arg: $!";
6462            ...
6463        }
6464    }
6465
6466Like the C<< <EXPR> >> operator, if a C<readline> expression is
6467used as the condition of a C<while> or C<for> loop, then it will be
6468implicitly assigned to C<$_>.  If either a C<readline> expression or
6469an explicit assignment of a C<readline> expression to a scalar is used
6470as a C<while>/C<for> condition, then the condition actually tests for
6471definedness of the expression's value, not for its regular truth value.
6472
6473=item readlink EXPR
6474X<readlink>
6475
6476=item readlink
6477
6478=for Pod::Functions determine where a symbolic link is pointing
6479
6480Returns the value of a symbolic link, if symbolic links are
6481implemented.  If not, raises an exception.  If there is a system
6482error, returns the undefined value and sets L<C<$!>|perlvar/$!> (errno).
6483If EXPR is omitted, uses L<C<$_>|perlvar/$_>.
6484
6485Portability issues: L<perlport/readlink>.
6486
6487=item readpipe EXPR
6488
6489=item readpipe
6490X<readpipe>
6491
6492=for Pod::Functions execute a system command and collect standard output
6493
6494EXPR is executed as a system command.
6495The collected standard output of the command is returned.
6496In scalar context, it comes back as a single (potentially
6497multi-line) string.  In list context, returns a list of lines
6498(however you've defined lines with L<C<$E<sol>>|perlvar/$E<sol>> (or
6499C<$INPUT_RECORD_SEPARATOR> in L<English>)).
6500This is the internal function implementing the C<qx/EXPR/>
6501operator, but you can use it directly.  The C<qx/EXPR/>
6502operator is discussed in more detail in L<perlop/"C<qx/I<STRING>/>">.
6503If EXPR is omitted, uses L<C<$_>|perlvar/$_>.
6504
6505=item recv SOCKET,SCALAR,LENGTH,FLAGS
6506X<recv>
6507
6508=for Pod::Functions receive a message over a Socket
6509
6510Receives a message on a socket.  Attempts to receive LENGTH characters
6511of data into variable SCALAR from the specified SOCKET filehandle.
6512SCALAR will be grown or shrunk to the length actually read.  Takes the
6513same flags as the system call of the same name.  Returns the address
6514of the sender if SOCKET's protocol supports this; returns an empty
6515string otherwise.  If there's an error, returns the undefined value.
6516This call is actually implemented in terms of the L<recvfrom(2)> system call.
6517See L<perlipc/"UDP: Message Passing"> for examples.
6518
6519Note that if the socket has been marked as C<:utf8>, C<recv> will
6520throw an exception.  The C<:encoding(...)> layer implicitly introduces
6521the C<:utf8> layer.  See L<C<binmode>|/binmode FILEHANDLE, LAYER>.
6522
6523=item redo LABEL
6524X<redo>
6525
6526=item redo EXPR
6527
6528=item redo
6529
6530=for Pod::Functions start this loop iteration over again
6531
6532The L<C<redo>|/redo LABEL> command restarts the loop block without
6533evaluating the conditional again.  The L<C<continue>|/continue BLOCK>
6534block, if any, is not executed.  If
6535the LABEL is omitted, the command refers to the innermost enclosing
6536loop.  The C<redo EXPR> form, available starting in Perl 5.18.0, allows a
6537label name to be computed at run time, and is otherwise identical to C<redo
6538LABEL>.  Programs that want to lie to themselves about what was just input
6539normally use this command:
6540
6541    # a simpleminded Pascal comment stripper
6542    # (warning: assumes no { or } in strings)
6543    LINE: while (<STDIN>) {
6544        while (s|({.*}.*){.*}|$1 |) {}
6545        s|{.*}| |;
6546        if (s|{.*| |) {
6547            my $front = $_;
6548            while (<STDIN>) {
6549                if (/}/) {  # end of comment?
6550                    s|^|$front\{|;
6551                    redo LINE;
6552                }
6553            }
6554        }
6555        print;
6556    }
6557
6558L<C<redo>|/redo LABEL> cannot return a value from a block that typically
6559returns a value, such as C<eval {}>, C<sub {}>, or C<do {}>. It will perform
6560its flow control behavior, which precludes any return value. It should not be
6561used to exit a L<C<grep>|/grep BLOCK LIST> or L<C<map>|/map BLOCK LIST>
6562operation.
6563
6564Note that a block by itself is semantically identical to a loop
6565that executes once.  Thus L<C<redo>|/redo LABEL> inside such a block
6566will effectively turn it into a looping construct.
6567
6568See also L<C<continue>|/continue BLOCK> for an illustration of how
6569L<C<last>|/last LABEL>, L<C<next>|/next LABEL>, and
6570L<C<redo>|/redo LABEL> work.
6571
6572Unlike most named operators, this has the same precedence as assignment.
6573It is also exempt from the looks-like-a-function rule, so
6574C<redo ("foo")."bar"> will cause "bar" to be part of the argument to
6575L<C<redo>|/redo LABEL>.
6576
6577=item ref EXPR
6578X<ref> X<reference>
6579
6580=item ref
6581
6582=for Pod::Functions find out the type of thing being referenced
6583
6584Examines the value of EXPR, expecting it to be a reference, and returns
6585a string giving information about the reference and the type of referent.
6586If EXPR is not specified, L<C<$_>|perlvar/$_> will be used.
6587
6588If the operand is not a reference, then the empty string will be returned.
6589An empty string will only be returned in this situation.  C<ref> is often
6590useful to just test whether a value is a reference, which can be done
6591by comparing the result to the empty string.  It is a common mistake
6592to use the result of C<ref> directly as a truth value: this goes wrong
6593because C<0> (which is false) can be returned for a reference.
6594
6595If the operand is a reference to a blessed object, then the name of
6596the class into which the referent is blessed will be returned.  C<ref>
6597doesn't care what the physical type of the referent is; blessing takes
6598precedence over such concerns.  Beware that exact comparison of C<ref>
6599results against a class name doesn't perform a class membership test:
6600a class's members also include objects blessed into subclasses, for
6601which C<ref> will return the name of the subclass.  Also beware that
6602class names can clash with the built-in type names (described below).
6603
6604If the operand is a reference to an unblessed object, then the return
6605value indicates the type of object.  If the unblessed referent is not
6606a scalar, then the return value will be one of the strings C<ARRAY>,
6607C<HASH>, C<CODE>, C<FORMAT>, or C<IO>, indicating only which kind of
6608object it is.  If the unblessed referent is a scalar, then the return
6609value will be one of the strings C<SCALAR>, C<VSTRING>, C<REF>, C<GLOB>,
6610C<LVALUE>, or C<REGEXP>, depending on the kind of value the scalar
6611currently has.   But note that C<qr//> scalars are created already
6612blessed, so C<ref qr/.../> will likely return C<Regexp>.  Beware that
6613these built-in type names can also be used as
6614class names, so C<ref> returning one of these names doesn't unambiguously
6615indicate that the referent is of the kind to which the name refers.
6616
6617The ambiguity between built-in type names and class names significantly
6618limits the utility of C<ref>.  For unambiguous information, use
6619L<C<Scalar::Util::blessed()>|Scalar::Util/blessed> for information about
6620blessing, and L<C<Scalar::Util::reftype()>|Scalar::Util/reftype> for
6621information about physical types.  Use L<the C<isa> method|UNIVERSAL/C<<
6622$obj->isa( TYPE ) >>> for class membership tests, though one must be
6623sure of blessedness before attempting a method call.  Alternatively, the
6624L<C<isa> operator|perlop/"Class Instance Operator"> can test class
6625membership without checking blessedness first.
6626
6627See also L<perlref> and L<perlobj>.
6628
6629=item rename OLDNAME,NEWNAME
6630X<rename> X<move> X<mv> X<ren>
6631
6632=for Pod::Functions change a filename
6633
6634Changes the name of a file; an existing file NEWNAME will be
6635clobbered.  Returns true for success; on failure returns false and sets
6636L<C<$!>|perlvar/$!>.
6637
6638Behavior of this function varies wildly depending on your system
6639implementation.  For example, it will usually not work across file system
6640boundaries, even though the system I<mv> command sometimes compensates
6641for this.  Other restrictions include whether it works on directories,
6642open files, or pre-existing files.  Check L<perlport> and either the
6643L<rename(2)> manpage or equivalent system documentation for details.
6644
6645For a platform independent L<C<move>|File::Copy/move> function look at
6646the L<File::Copy> module.
6647
6648Portability issues: L<perlport/rename>.
6649
6650=item require VERSION
6651X<require>
6652
6653=item require EXPR
6654
6655=item require
6656
6657=for Pod::Functions load in external functions from a library at runtime
6658
6659Demands a version of Perl specified by VERSION, or demands some semantics
6660specified by EXPR or by L<C<$_>|perlvar/$_> if EXPR is not supplied.
6661
6662VERSION may be either a literal such as v5.24.1, which will be
6663compared to L<C<$^V>|perlvar/$^V> (or C<$PERL_VERSION> in L<English>),
6664or a numeric argument of the form 5.024001, which will be compared to
6665L<C<$]>|perlvar/$]>. An exception is raised if VERSION is greater than
6666the version of the current Perl interpreter.  Compare with
6667L<C<use>|/use Module VERSION LIST>, which can do a similar check at
6668compile time.
6669
6670Specifying VERSION as a numeric argument of the form 5.024001 should
6671generally be avoided as older less readable syntax compared to
6672v5.24.1. Before perl 5.8.0 (released in 2002), the more verbose numeric
6673form was the only supported syntax, which is why you might see it in
6674older code.
6675
6676    require v5.24.1;    # run time version check
6677    require 5.24.1;     # ditto
6678    require 5.024_001;  # ditto; older syntax compatible
6679                          with perl 5.6
6680
6681Otherwise, L<C<require>|/require VERSION> demands that a library file be
6682included if it hasn't already been included.  The file is included via
6683the do-FILE mechanism, which is essentially just a variety of
6684L<C<eval>|/eval EXPR> with the
6685caveat that lexical variables in the invoking script will be invisible
6686to the included code.  If it were implemented in pure Perl, it
6687would have semantics similar to the following:
6688
6689    use Carp 'croak';
6690    use version;
6691
6692    sub require {
6693        my ($filename) = @_;
6694        if ( my $version = eval { version->parse($filename) } ) {
6695            if ( $version > $^V ) {
6696               my $vn = $version->normal;
6697               croak "Perl $vn required--this is only $^V, stopped";
6698            }
6699            return 1;
6700        }
6701
6702        if (exists $INC{$filename}) {
6703            return 1 if $INC{$filename};
6704            croak "Compilation failed in require";
6705        }
6706
6707        foreach $prefix (@INC) {
6708            if (ref($prefix)) {
6709                #... do other stuff - see text below ....
6710            }
6711            # (see text below about possible appending of .pmc
6712            # suffix to $filename)
6713            my $realfilename = "$prefix/$filename";
6714            next if ! -e $realfilename || -d _ || -b _;
6715            $INC{$filename} = $realfilename;
6716            my $result = do($realfilename);
6717                         # but run in caller's namespace
6718
6719            if (!defined $result) {
6720                $INC{$filename} = undef;
6721                croak $@ ? "$@Compilation failed in require"
6722                         : "Can't locate $filename: $!\n";
6723            }
6724            if (!$result) {
6725                delete $INC{$filename};
6726                croak "$filename did not return true value";
6727            }
6728            $! = 0;
6729            return $result;
6730        }
6731        croak "Can't locate $filename in \@INC ...";
6732    }
6733
6734Note that the file will not be included twice under the same specified
6735name.
6736
6737The file must return true as the last statement to indicate
6738successful execution of any initialization code, so it's customary to
6739end such a file with C<1;> unless you're sure it'll return true
6740otherwise.  But it's better just to put the C<1;>, in case you add more
6741statements.
6742
6743If EXPR is a bareword, L<C<require>|/require VERSION> assumes a F<.pm>
6744extension and replaces C<::> with C</> in the filename for you,
6745to make it easy to load standard modules.  This form of loading of
6746modules does not risk altering your namespace, however it will autovivify
6747the stash for the required module.
6748
6749In other words, if you try this:
6750
6751        require Foo::Bar;     # a splendid bareword
6752
6753The require function will actually look for the F<Foo/Bar.pm> file in the
6754directories specified in the L<C<@INC>|perlvar/@INC> array, and it will
6755autovivify the C<Foo::Bar::> stash at compile time.
6756
6757But if you try this:
6758
6759        my $class = 'Foo::Bar';
6760        require $class;       # $class is not a bareword
6761    #or
6762        require "Foo::Bar";   # not a bareword because of the ""
6763
6764The require function will look for the F<Foo::Bar> file in the
6765L<C<@INC>|perlvar/@INC>  array and
6766will complain about not finding F<Foo::Bar> there.  In this case you can do:
6767
6768        eval "require $class";
6769
6770or you could do
6771
6772        require "Foo/Bar.pm";
6773
6774Neither of these forms will autovivify any stashes at compile time and
6775only have run time effects.
6776
6777Now that you understand how L<C<require>|/require VERSION> looks for
6778files with a bareword argument, there is a little extra functionality
6779going on behind the scenes.  Before L<C<require>|/require VERSION> looks
6780for a F<.pm> extension, it will first look for a similar filename with a
6781F<.pmc> extension.  If this file is found, it will be loaded in place of
6782any file ending in a F<.pm> extension. This applies to both the explicit
6783C<require "Foo/Bar.pm";> form and the C<require Foo::Bar;> form.
6784
6785You can also insert hooks into the import facility by putting Perl code
6786directly into the L<C<@INC>|perlvar/@INC> array.  There are three forms
6787of hooks: subroutine references, array references, and blessed objects.
6788
6789Subroutine references are the simplest case.  When the inclusion system
6790walks through L<C<@INC>|perlvar/@INC> and encounters a subroutine, this
6791subroutine gets called with two parameters, the first a reference to
6792itself, and the second the name of the file to be included (e.g.,
6793F<Foo/Bar.pm>).  The subroutine should return either nothing or else a
6794list of up to four values in the following order:
6795
6796=over
6797
6798=item 1
6799
6800A reference to a scalar, containing any initial source code to prepend to
6801the file or generator output.
6802
6803=item 2
6804
6805A filehandle, from which the file will be read.
6806
6807=item 3
6808
6809A reference to a subroutine.  If there is no filehandle (previous item),
6810then this subroutine is expected to generate one line of source code per
6811call, writing the line into L<C<$_>|perlvar/$_> and returning 1, then
6812finally at end of file returning 0.  If there is a filehandle, then the
6813subroutine will be called to act as a simple source filter, with the
6814line as read in L<C<$_>|perlvar/$_>.
6815Again, return 1 for each valid line, and 0 after all lines have been
6816returned.
6817For historical reasons the subroutine will receive a meaningless argument
6818(in fact always the numeric value zero) as C<$_[0]>.
6819
6820=item 4
6821
6822Optional state for the subroutine.  The state is passed in as C<$_[1]>.
6823
6824=back
6825
6826If an empty list, L<C<undef>|/undef EXPR>, or nothing that matches the
6827first 3 values above is returned, then L<C<require>|/require VERSION>
6828looks at the remaining elements of L<C<@INC>|perlvar/@INC>.
6829Note that this filehandle must be a real filehandle (strictly a typeglob
6830or reference to a typeglob, whether blessed or unblessed); tied filehandles
6831will be ignored and processing will stop there.
6832
6833If the hook is an array reference, its first element must be a subroutine
6834reference.  This subroutine is called as above, but the first parameter is
6835the array reference.  This lets you indirectly pass arguments to
6836the subroutine.
6837
6838In other words, you can write:
6839
6840    push @INC, \&my_sub;
6841    sub my_sub {
6842        my ($coderef, $filename) = @_;  # $coderef is \&my_sub
6843        ...
6844    }
6845
6846or:
6847
6848    push @INC, [ \&my_sub, $x, $y, ... ];
6849    sub my_sub {
6850        my ($arrayref, $filename) = @_;
6851        # Retrieve $x, $y, ...
6852        my (undef, @parameters) = @$arrayref;
6853        ...
6854    }
6855
6856If the hook is an object, it must provide an C<INC> method that will be
6857called as above, the first parameter being the object itself.  (Note that
6858you must fully qualify the sub's name, as unqualified C<INC> is always forced
6859into package C<main>.)  Here is a typical code layout:
6860
6861    # In Foo.pm
6862    package Foo;
6863    sub new { ... }
6864    sub Foo::INC {
6865        my ($self, $filename) = @_;
6866        ...
6867    }
6868
6869    # In the main program
6870    push @INC, Foo->new(...);
6871
6872These hooks are also permitted to set the L<C<%INC>|perlvar/%INC> entry
6873corresponding to the files they have loaded.  See L<perlvar/%INC>.
6874
6875For a yet-more-powerful import facility, see
6876L<C<use>|/use Module VERSION LIST> and L<perlmod>.
6877
6878=item reset EXPR
6879X<reset>
6880
6881=item reset
6882
6883=for Pod::Functions clear all variables of a given name
6884
6885Generally used in a L<C<continue>|/continue BLOCK> block at the end of a
6886loop to clear variables and reset C<m?pattern?> searches so that they
6887work again.  The
6888expression is interpreted as a list of single characters (hyphens
6889allowed for ranges).  All variables (scalars, arrays, and hashes)
6890in the current package beginning with one of
6891those letters are reset to their pristine state.  If the expression is
6892omitted, one-match searches (C<m?pattern?>) are reset to match again.
6893Only resets variables or searches in the current package.  Always returns
68941.  Examples:
6895
6896    reset 'X';      # reset all X variables
6897    reset 'a-z';    # reset lower case variables
6898    reset;          # just reset m?one-time? searches
6899
6900Resetting C<"A-Z"> is not recommended because you'll wipe out your
6901L<C<@ARGV>|perlvar/@ARGV> and L<C<@INC>|perlvar/@INC> arrays and your
6902L<C<%ENV>|perlvar/%ENV> hash.
6903
6904Resets only package variables; lexical variables are unaffected, but
6905they clean themselves up on scope exit anyway, so you'll probably want
6906to use them instead.  See L<C<my>|/my VARLIST>.
6907
6908=item return EXPR
6909X<return>
6910
6911=item return
6912
6913=for Pod::Functions get out of a function early
6914
6915Returns from a subroutine, L<C<eval>|/eval EXPR>,
6916L<C<do FILE>|/do EXPR>, L<C<sort>|/sort SUBNAME LIST> block or regex
6917eval block (but not a L<C<grep>|/grep BLOCK LIST>,
6918L<C<map>|/map BLOCK LIST>, or L<C<do BLOCK>|/do BLOCK> block) with the value
6919given in EXPR.  Evaluation of EXPR may be in list, scalar, or void
6920context, depending on how the return value will be used, and the context
6921may vary from one execution to the next (see
6922L<C<wantarray>|/wantarray>).  If no EXPR
6923is given, returns an empty list in list context, the undefined value in
6924scalar context, and (of course) nothing at all in void context.
6925
6926(In the absence of an explicit L<C<return>|/return EXPR>, a subroutine,
6927L<C<eval>|/eval EXPR>,
6928or L<C<do FILE>|/do EXPR> automatically returns the value of the last expression
6929evaluated.)
6930
6931Unlike most named operators, this is also exempt from the
6932looks-like-a-function rule, so C<return ("foo")."bar"> will
6933cause C<"bar"> to be part of the argument to L<C<return>|/return EXPR>.
6934
6935=item reverse LIST
6936X<reverse> X<rev> X<invert>
6937
6938=for Pod::Functions flip a string or a list
6939
6940In list context, returns a list value consisting of the elements
6941of LIST in the opposite order.  In scalar context, concatenates the
6942elements of LIST and returns a string value with all characters
6943in the opposite order.
6944
6945    print join(", ", reverse "world", "Hello"); # Hello, world
6946
6947    print scalar reverse "dlrow ,", "olleH";    # Hello, world
6948
6949Used without arguments in scalar context, L<C<reverse>|/reverse LIST>
6950reverses L<C<$_>|perlvar/$_>.
6951
6952    $_ = "dlrow ,olleH";
6953    print reverse;                         # No output, list context
6954    print scalar reverse;                  # Hello, world
6955
6956Note that reversing an array to itself (as in C<@a = reverse @a>) will
6957preserve non-existent elements whenever possible; i.e., for non-magical
6958arrays or for tied arrays with C<EXISTS> and C<DELETE> methods.
6959
6960This operator is also handy for inverting a hash, although there are some
6961caveats.  If a value is duplicated in the original hash, only one of those
6962can be represented as a key in the inverted hash.  Also, this has to
6963unwind one hash and build a whole new one, which may take some time
6964on a large hash, such as from a DBM file.
6965
6966    my %by_name = reverse %by_address;  # Invert the hash
6967
6968=item rewinddir DIRHANDLE
6969X<rewinddir>
6970
6971=for Pod::Functions reset directory handle
6972
6973Sets the current position to the beginning of the directory for the
6974L<C<readdir>|/readdir DIRHANDLE> routine on DIRHANDLE.
6975
6976Portability issues: L<perlport/rewinddir>.
6977
6978=item rindex STR,SUBSTR,POSITION
6979X<rindex>
6980
6981=item rindex STR,SUBSTR
6982
6983=for Pod::Functions right-to-left substring search
6984
6985Works just like L<C<index>|/index STR,SUBSTR,POSITION> except that it
6986returns the position of the I<last>
6987occurrence of SUBSTR in STR.  If POSITION is specified, returns the
6988last occurrence beginning at or before that position.
6989
6990=item rmdir FILENAME
6991X<rmdir> X<rd> X<directory, remove>
6992
6993=item rmdir
6994
6995=for Pod::Functions remove a directory
6996
6997Deletes the directory specified by FILENAME if that directory is
6998empty.  If it succeeds it returns true; otherwise it returns false and
6999sets L<C<$!>|perlvar/$!> (errno).  If FILENAME is omitted, uses
7000L<C<$_>|perlvar/$_>.
7001
7002To remove a directory tree recursively (C<rm -rf> on Unix) look at
7003the L<C<rmtree>|File::Path/rmtree( $dir )> function of the L<File::Path>
7004module.
7005
7006=item s///
7007
7008=for Pod::Functions replace a pattern with a string
7009
7010The substitution operator.  See L<perlop/"Regexp Quote-Like Operators">.
7011
7012=item say FILEHANDLE LIST
7013X<say>
7014
7015=item say FILEHANDLE
7016
7017=item say LIST
7018
7019=item say
7020
7021=for Pod::Functions +say output a list to a filehandle, appending a newline
7022
7023Just like L<C<print>|/print FILEHANDLE LIST>, but implicitly appends a
7024newline at the end of the LIST instead of any value L<C<$\>|perlvar/$\>
7025might have.  To use FILEHANDLE without a LIST to
7026print the contents of L<C<$_>|perlvar/$_> to it, you must use a bareword
7027filehandle like C<FH>, not an indirect one like C<$fh>.
7028
7029L<C<say>|/say FILEHANDLE LIST> is available only if the
7030L<C<"say"> feature|feature/The 'say' feature> is enabled or if it is
7031prefixed with C<CORE::>.  The
7032L<C<"say"> feature|feature/The 'say' feature> is enabled automatically
7033with a C<use v5.10> (or higher) declaration in the current scope.
7034
7035=item scalar EXPR
7036X<scalar> X<context>
7037
7038=for Pod::Functions force a scalar context
7039
7040Forces EXPR to be interpreted in scalar context and returns the value
7041of EXPR.
7042
7043    my @counts = ( scalar @a, scalar @b, scalar @c );
7044
7045There is no equivalent operator to force an expression to
7046be interpolated in list context because in practice, this is never
7047needed.  If you really wanted to do so, however, you could use
7048the construction C<@{[ (some expression) ]}>, but usually a simple
7049C<(some expression)> suffices.
7050
7051Because L<C<scalar>|/scalar EXPR> is a unary operator, if you
7052accidentally use a
7053parenthesized list for the EXPR, this behaves as a scalar comma expression,
7054evaluating all but the last element in void context and returning the final
7055element evaluated in scalar context.  This is seldom what you want.
7056
7057The following single statement:
7058
7059    print uc(scalar(foo(), $bar)), $baz;
7060
7061is the moral equivalent of these two:
7062
7063    foo();
7064    print(uc($bar), $baz);
7065
7066See L<perlop> for more details on unary operators and the comma operator,
7067and L<perldata> for details on evaluating a hash in scalar context.
7068
7069=item seek FILEHANDLE,POSITION,WHENCE
7070X<seek> X<fseek> X<filehandle, position>
7071
7072=for Pod::Functions reposition file pointer for random-access I/O
7073
7074Sets FILEHANDLE's position, just like the L<fseek(3)> call of C C<stdio>.
7075FILEHANDLE may be an expression whose value gives the name of the
7076filehandle.  The values for WHENCE are C<0> to set the new position
7077I<in bytes> to POSITION; C<1> to set it to the current position plus
7078POSITION; and C<2> to set it to EOF plus POSITION, typically
7079negative.  For WHENCE you may use the constants C<SEEK_SET>,
7080C<SEEK_CUR>, and C<SEEK_END> (start of the file, current position, end
7081of the file) from the L<Fcntl> module.  Returns C<1> on success, false
7082otherwise.
7083
7084Note the emphasis on bytes: even if the filehandle has been set to operate
7085on characters (for example using the C<:encoding(UTF-8)> I/O layer), the
7086L<C<seek>|/seek FILEHANDLE,POSITION,WHENCE>,
7087L<C<tell>|/tell FILEHANDLE>, and
7088L<C<sysseek>|/sysseek FILEHANDLE,POSITION,WHENCE>
7089family of functions use byte offsets, not character offsets,
7090because seeking to a character offset would be very slow in a UTF-8 file.
7091
7092If you want to position the file for
7093L<C<sysread>|/sysread FILEHANDLE,SCALAR,LENGTH,OFFSET> or
7094L<C<syswrite>|/syswrite FILEHANDLE,SCALAR,LENGTH,OFFSET>, don't use
7095L<C<seek>|/seek FILEHANDLE,POSITION,WHENCE>, because buffering makes its
7096effect on the file's read-write position unpredictable and non-portable.
7097Use L<C<sysseek>|/sysseek FILEHANDLE,POSITION,WHENCE> instead.
7098
7099Due to the rules and rigors of ANSI C, on some systems you have to do a
7100seek whenever you switch between reading and writing.  Amongst other
7101things, this may have the effect of calling stdio's L<clearerr(3)>.
7102A WHENCE of C<1> (C<SEEK_CUR>) is useful for not moving the file position:
7103
7104    seek($fh, 0, 1);
7105
7106This is also useful for applications emulating C<tail -f>.  Once you hit
7107EOF on your read and then sleep for a while, you (probably) have to stick in a
7108dummy L<C<seek>|/seek FILEHANDLE,POSITION,WHENCE> to reset things.  The
7109L<C<seek>|/seek FILEHANDLE,POSITION,WHENCE> doesn't change the position,
7110but it I<does> clear the end-of-file condition on the handle, so that the
7111next C<readline FILE> makes Perl try again to read something.  (We hope.)
7112
7113If that doesn't work (some I/O implementations are particularly
7114cantankerous), you might need something like this:
7115
7116    for (;;) {
7117        for ($curpos = tell($fh); $_ = readline($fh);
7118             $curpos = tell($fh)) {
7119            # search for some stuff and put it into files
7120        }
7121        sleep($for_a_while);
7122        seek($fh, $curpos, 0);
7123    }
7124
7125=item seekdir DIRHANDLE,POS
7126X<seekdir>
7127
7128=for Pod::Functions reposition directory pointer
7129
7130Sets the current position for the L<C<readdir>|/readdir DIRHANDLE>
7131routine on DIRHANDLE.  POS must be a value returned by
7132L<C<telldir>|/telldir DIRHANDLE>.  L<C<seekdir>|/seekdir DIRHANDLE,POS>
7133also has the same caveats about possible directory compaction as the
7134corresponding system library routine.
7135
7136=item select FILEHANDLE
7137X<select> X<filehandle, default>
7138
7139=item select
7140
7141=for Pod::Functions reset default output or do I/O multiplexing
7142
7143Returns the currently selected filehandle.  If FILEHANDLE is supplied,
7144sets the new current default filehandle for output.  This has two
7145effects: first, a L<C<write>|/write FILEHANDLE> or a L<C<print>|/print
7146FILEHANDLE LIST> without a filehandle
7147default to this FILEHANDLE.  Second, references to variables related to
7148output will refer to this output channel.
7149
7150For example, to set the top-of-form format for more than one
7151output channel, you might do the following:
7152
7153    select(REPORT1);
7154    $^ = 'report1_top';
7155    select(REPORT2);
7156    $^ = 'report2_top';
7157
7158FILEHANDLE may be an expression whose value gives the name of the
7159actual filehandle.  Thus:
7160
7161    my $oldfh = select(STDERR); $| = 1; select($oldfh);
7162
7163Some programmers may prefer to think of filehandles as objects with
7164methods, preferring to write the last example as:
7165
7166    STDERR->autoflush(1);
7167
7168(Prior to Perl version 5.14, you have to C<use IO::Handle;> explicitly
7169first.)
7170
7171Portability issues: L<perlport/select>.
7172
7173=item select RBITS,WBITS,EBITS,TIMEOUT
7174X<select>
7175
7176This calls the L<select(2)> syscall with the bit masks specified, which
7177can be constructed using L<C<fileno>|/fileno FILEHANDLE> and
7178L<C<vec>|/vec EXPR,OFFSET,BITS>, along these lines:
7179
7180    my $rin = my $win = my $ein = '';
7181    vec($rin, fileno(STDIN),  1) = 1;
7182    vec($win, fileno(STDOUT), 1) = 1;
7183    $ein = $rin | $win;
7184
7185If you want to select on many filehandles, you may wish to write a
7186subroutine like this:
7187
7188    sub fhbits {
7189        my @fhlist = @_;
7190        my $bits = "";
7191        for my $fh (@fhlist) {
7192            vec($bits, fileno($fh), 1) = 1;
7193        }
7194        return $bits;
7195    }
7196    my $rin = fhbits(\*STDIN, $tty, $mysock);
7197
7198The usual idiom is:
7199
7200 my ($nfound, $timeleft) =
7201   select(my $rout = $rin, my $wout = $win, my $eout = $ein,
7202                                                          $timeout);
7203
7204or to block until something becomes ready just do this
7205
7206 my $nfound =
7207   select(my $rout = $rin, my $wout = $win, my $eout = $ein, undef);
7208
7209Most systems do not bother to return anything useful in C<$timeleft>, so
7210calling L<C<select>|/select RBITS,WBITS,EBITS,TIMEOUT> in scalar context
7211just returns C<$nfound>.
7212
7213Any of the bit masks can also be L<C<undef>|/undef EXPR>.  The timeout,
7214if specified, is
7215in seconds, which may be fractional.  Note: not all implementations are
7216capable of returning the C<$timeleft>.  If not, they always return
7217C<$timeleft> equal to the supplied C<$timeout>.
7218
7219You can effect a sleep of 250 milliseconds this way:
7220
7221    select(undef, undef, undef, 0.25);
7222
7223Note that whether L<C<select>|/select RBITS,WBITS,EBITS,TIMEOUT> gets
7224restarted after signals (say, SIGALRM) is implementation-dependent.  See
7225also L<perlport> for notes on the portability of
7226L<C<select>|/select RBITS,WBITS,EBITS,TIMEOUT>.
7227
7228On error, L<C<select>|/select RBITS,WBITS,EBITS,TIMEOUT> behaves just
7229like L<select(2)>: it returns C<-1> and sets L<C<$!>|perlvar/$!>.
7230
7231On some Unixes, L<select(2)> may report a socket file descriptor as
7232"ready for reading" even when no data is available, and thus any
7233subsequent L<C<read>|/read FILEHANDLE,SCALAR,LENGTH,OFFSET> would block.
7234This can be avoided if you always use C<O_NONBLOCK> on the socket.  See
7235L<select(2)> and L<fcntl(2)> for further details.
7236
7237The standard L<C<IO::Select>|IO::Select> module provides a
7238user-friendlier interface to
7239L<C<select>|/select RBITS,WBITS,EBITS,TIMEOUT>, mostly because it does
7240all the bit-mask work for you.
7241
7242B<WARNING>: One should not attempt to mix buffered I/O (like
7243L<C<read>|/read FILEHANDLE,SCALAR,LENGTH,OFFSET> or
7244L<C<readline>|/readline EXPR>) with
7245L<C<select>|/select RBITS,WBITS,EBITS,TIMEOUT>, except as permitted by
7246POSIX, and even then only on POSIX systems.  You have to use
7247L<C<sysread>|/sysread FILEHANDLE,SCALAR,LENGTH,OFFSET> instead.
7248
7249Portability issues: L<perlport/select>.
7250
7251=item semctl ID,SEMNUM,CMD,ARG
7252X<semctl>
7253
7254=for Pod::Functions SysV semaphore control operations
7255
7256Calls the System V IPC function L<semctl(2)>.  You'll probably have to say
7257
7258    use IPC::SysV;
7259
7260first to get the correct constant definitions.  If CMD is IPC_STAT or
7261GETALL, then ARG must be a variable that will hold the returned
7262semid_ds structure or semaphore value array.  Returns like
7263L<C<ioctl>|/ioctl FILEHANDLE,FUNCTION,SCALAR>:
7264the undefined value for error, "C<0 but true>" for zero, or the actual
7265return value otherwise.  The ARG must consist of a vector of native
7266short integers, which may be created with C<pack("s!",(0)x$nsem)>.
7267See also L<perlipc/"SysV IPC"> and the documentation for
7268L<C<IPC::SysV>|IPC::SysV> and L<C<IPC::Semaphore>|IPC::Semaphore>.
7269
7270Portability issues: L<perlport/semctl>.
7271
7272=item semget KEY,NSEMS,FLAGS
7273X<semget>
7274
7275=for Pod::Functions get set of SysV semaphores
7276
7277Calls the System V IPC function L<semget(2)>.  Returns the semaphore id, or
7278the undefined value on error.  See also
7279L<perlipc/"SysV IPC"> and the documentation for
7280L<C<IPC::SysV>|IPC::SysV> and L<C<IPC::Semaphore>|IPC::Semaphore>.
7281
7282Portability issues: L<perlport/semget>.
7283
7284=item semop KEY,OPSTRING
7285X<semop>
7286
7287=for Pod::Functions SysV semaphore operations
7288
7289Calls the System V IPC function L<semop(2)> for semaphore operations
7290such as signalling and waiting.  OPSTRING must be a packed array of
7291semop structures.  Each semop structure can be generated with
7292C<pack("s!3", $semnum, $semop, $semflag)>.  The length of OPSTRING
7293implies the number of semaphore operations.  Returns true if
7294successful, false on error.  As an example, the
7295following code waits on semaphore $semnum of semaphore id $semid:
7296
7297    my $semop = pack("s!3", $semnum, -1, 0);
7298    die "Semaphore trouble: $!\n" unless semop($semid, $semop);
7299
7300To signal the semaphore, replace C<-1> with C<1>.  See also
7301L<perlipc/"SysV IPC"> and the documentation for
7302L<C<IPC::SysV>|IPC::SysV> and L<C<IPC::Semaphore>|IPC::Semaphore>.
7303
7304Portability issues: L<perlport/semop>.
7305
7306=item send SOCKET,MSG,FLAGS,TO
7307X<send>
7308
7309=item send SOCKET,MSG,FLAGS
7310
7311=for Pod::Functions send a message over a socket
7312
7313Sends a message on a socket.  Attempts to send the scalar MSG to the SOCKET
7314filehandle.  Takes the same flags as the system call of the same name.  On
7315unconnected sockets, you must specify a destination to I<send to>, in which
7316case it does a L<sendto(2)> syscall.  Returns the number of characters sent,
7317or the undefined value on error.  The L<sendmsg(2)> syscall is currently
7318unimplemented.  See L<perlipc/"UDP: Message Passing"> for examples.
7319
7320Note that if the socket has been marked as C<:utf8>, C<send> will
7321throw an exception.  The C<:encoding(...)> layer implicitly introduces
7322the C<:utf8> layer.  See L<C<binmode>|/binmode FILEHANDLE, LAYER>.
7323
7324=item setpgrp PID,PGRP
7325X<setpgrp> X<group>
7326
7327=for Pod::Functions set the process group of a process
7328
7329Sets the current process group for the specified PID, C<0> for the current
7330process.  Raises an exception when used on a machine that doesn't
7331implement POSIX L<setpgid(2)> or BSD L<setpgrp(2)>.  If the arguments
7332are omitted, it defaults to C<0,0>.  Note that the BSD 4.2 version of
7333L<C<setpgrp>|/setpgrp PID,PGRP> does not accept any arguments, so only
7334C<setpgrp(0,0)> is portable.  See also
7335L<C<POSIX::setsid()>|POSIX/C<setsid>>.
7336
7337Portability issues: L<perlport/setpgrp>.
7338
7339=item setpriority WHICH,WHO,PRIORITY
7340X<setpriority> X<priority> X<nice> X<renice>
7341
7342=for Pod::Functions set a process's nice value
7343
7344Sets the current priority for a process, a process group, or a user.
7345(See L<setpriority(2)>.)  Raises an exception when used on a machine
7346that doesn't implement L<setpriority(2)>.
7347
7348C<WHICH> can be any of C<PRIO_PROCESS>, C<PRIO_PGRP> or C<PRIO_USER>
7349imported from L<POSIX/RESOURCE CONSTANTS>.
7350
7351Portability issues: L<perlport/setpriority>.
7352
7353=item setsockopt SOCKET,LEVEL,OPTNAME,OPTVAL
7354X<setsockopt>
7355
7356=for Pod::Functions set some socket options
7357
7358Sets the socket option requested.  Returns L<C<undef>|/undef EXPR> on
7359error.  Use integer constants provided by the L<C<Socket>|Socket> module
7360for
7361LEVEL and OPNAME.  Values for LEVEL can also be obtained from
7362getprotobyname.  OPTVAL might either be a packed string or an integer.
7363An integer OPTVAL is shorthand for pack("i", OPTVAL).
7364
7365An example disabling Nagle's algorithm on a socket:
7366
7367    use Socket qw(IPPROTO_TCP TCP_NODELAY);
7368    setsockopt($socket, IPPROTO_TCP, TCP_NODELAY, 1);
7369
7370Portability issues: L<perlport/setsockopt>.
7371
7372=item shift ARRAY
7373X<shift>
7374
7375=item shift
7376
7377=for Pod::Functions remove the first element of an array, and return it
7378
7379Shifts the first value of the array off and returns it, shortening the
7380array by 1 and moving everything down.  If there are no elements in the
7381array, returns the undefined value.  If ARRAY is omitted, shifts the
7382L<C<@_>|perlvar/@_> array within the lexical scope of subroutines and
7383formats, and the L<C<@ARGV>|perlvar/@ARGV> array outside a subroutine
7384and also within the lexical scopes
7385established by the C<eval STRING>, C<BEGIN {}>, C<INIT {}>, C<CHECK {}>,
7386C<UNITCHECK {}>, and C<END {}> constructs.
7387
7388Starting with Perl 5.14, an experimental feature allowed
7389L<C<shift>|/shift ARRAY> to take a
7390scalar expression. This experiment has been deemed unsuccessful, and was
7391removed as of Perl 5.24.
7392
7393See also L<C<unshift>|/unshift ARRAY,LIST>, L<C<push>|/push ARRAY,LIST>,
7394and L<C<pop>|/pop ARRAY>.  L<C<shift>|/shift ARRAY> and
7395L<C<unshift>|/unshift ARRAY,LIST> do the same thing to the left end of
7396an array that L<C<pop>|/pop ARRAY> and L<C<push>|/push ARRAY,LIST> do to
7397the right end.
7398
7399=item shmctl ID,CMD,ARG
7400X<shmctl>
7401
7402=for Pod::Functions SysV shared memory operations
7403
7404Calls the System V IPC function shmctl.  You'll probably have to say
7405
7406    use IPC::SysV;
7407
7408first to get the correct constant definitions.  If CMD is C<IPC_STAT>,
7409then ARG must be a variable that will hold the returned C<shmid_ds>
7410structure.  Returns like ioctl: L<C<undef>|/undef EXPR> for error; "C<0>
7411but true" for zero; and the actual return value otherwise.
7412See also L<perlipc/"SysV IPC"> and the documentation for
7413L<C<IPC::SysV>|IPC::SysV>.
7414
7415Portability issues: L<perlport/shmctl>.
7416
7417=item shmget KEY,SIZE,FLAGS
7418X<shmget>
7419
7420=for Pod::Functions get SysV shared memory segment identifier
7421
7422Calls the System V IPC function shmget.  Returns the shared memory
7423segment id, or L<C<undef>|/undef EXPR> on error.
7424See also L<perlipc/"SysV IPC"> and the documentation for
7425L<C<IPC::SysV>|IPC::SysV>.
7426
7427Portability issues: L<perlport/shmget>.
7428
7429=item shmread ID,VAR,POS,SIZE
7430X<shmread>
7431X<shmwrite>
7432
7433=for Pod::Functions read SysV shared memory
7434
7435=item shmwrite ID,STRING,POS,SIZE
7436
7437=for Pod::Functions write SysV shared memory
7438
7439Reads or writes the System V shared memory segment ID starting at
7440position POS for size SIZE by attaching to it, copying in/out, and
7441detaching from it.  When reading, VAR must be a variable that will
7442hold the data read.  When writing, if STRING is too long, only SIZE
7443bytes are used; if STRING is too short, nulls are written to fill out
7444SIZE bytes.  Return true if successful, false on error.
7445L<C<shmread>|/shmread ID,VAR,POS,SIZE> taints the variable.  See also
7446L<perlipc/"SysV IPC"> and the documentation for
7447L<C<IPC::SysV>|IPC::SysV> and the L<C<IPC::Shareable>|IPC::Shareable>
7448module from CPAN.
7449
7450Portability issues: L<perlport/shmread> and L<perlport/shmwrite>.
7451
7452=item shutdown SOCKET,HOW
7453X<shutdown>
7454
7455=for Pod::Functions close down just half of a socket connection
7456
7457Shuts down a socket connection in the manner indicated by HOW, which
7458has the same interpretation as in the syscall of the same name.
7459
7460    shutdown($socket, 0);    # I/we have stopped reading data
7461    shutdown($socket, 1);    # I/we have stopped writing data
7462    shutdown($socket, 2);    # I/we have stopped using this socket
7463
7464This is useful with sockets when you want to tell the other
7465side you're done writing but not done reading, or vice versa.
7466It's also a more insistent form of close because it also
7467disables the file descriptor in any forked copies in other
7468processes.
7469
7470Returns C<1> for success; on error, returns L<C<undef>|/undef EXPR> if
7471the first argument is not a valid filehandle, or returns C<0> and sets
7472L<C<$!>|perlvar/$!> for any other failure.
7473
7474=item sin EXPR
7475X<sin> X<sine> X<asin> X<arcsine>
7476
7477=item sin
7478
7479=for Pod::Functions return the sine of a number
7480
7481Returns the sine of EXPR (expressed in radians).  If EXPR is omitted,
7482returns sine of L<C<$_>|perlvar/$_>.
7483
7484For the inverse sine operation, you may use the C<Math::Trig::asin>
7485function, or use this relation:
7486
7487    sub asin { atan2($_[0], sqrt(1 - $_[0] * $_[0])) }
7488
7489=item sleep EXPR
7490X<sleep> X<pause>
7491
7492=item sleep
7493
7494=for Pod::Functions block for some number of seconds
7495
7496Causes the script to sleep for (integer) EXPR seconds, or forever if no
7497argument is given.  Returns the integer number of seconds actually slept.
7498
7499EXPR should be a positive integer. If called with a negative integer,
7500L<C<sleep>|/sleep EXPR> does not sleep but instead emits a warning, sets
7501$! (C<errno>), and returns zero.
7502
7503C<sleep 0> is permitted, but a function call to the underlying platform
7504implementation still occurs, with any side effects that may have.
7505C<sleep 0> is therefore not exactly identical to not sleeping at all.
7506
7507May be interrupted if the process receives a signal such as C<SIGALRM>.
7508
7509    eval {
7510        local $SIG{ALRM} = sub { die "Alarm!\n" };
7511        sleep;
7512    };
7513    die $@ unless $@ eq "Alarm!\n";
7514
7515You probably cannot mix L<C<alarm>|/alarm SECONDS> and
7516L<C<sleep>|/sleep EXPR> calls, because L<C<sleep>|/sleep EXPR> is often
7517implemented using L<C<alarm>|/alarm SECONDS>.
7518
7519On some older systems, it may sleep up to a full second less than what
7520you requested, depending on how it counts seconds.  Most modern systems
7521always sleep the full amount.  They may appear to sleep longer than that,
7522however, because your process might not be scheduled right away in a
7523busy multitasking system.
7524
7525For delays of finer granularity than one second, the L<Time::HiRes>
7526module (from CPAN, and starting from Perl 5.8 part of the standard
7527distribution) provides L<C<usleep>|Time::HiRes/usleep ( $useconds )>.
7528You may also use Perl's four-argument
7529version of L<C<select>|/select RBITS,WBITS,EBITS,TIMEOUT> leaving the
7530first three arguments undefined, or you might be able to use the
7531L<C<syscall>|/syscall NUMBER, LIST> interface to access L<setitimer(2)>
7532if your system supports it.  See L<perlfaq8> for details.
7533
7534See also the L<POSIX> module's L<C<pause>|POSIX/C<pause>> function.
7535
7536=item socket SOCKET,DOMAIN,TYPE,PROTOCOL
7537X<socket>
7538
7539=for Pod::Functions create a socket
7540
7541Opens a socket of the specified kind and attaches it to filehandle
7542SOCKET.  DOMAIN, TYPE, and PROTOCOL are specified the same as for
7543the syscall of the same name.  You should C<use Socket> first
7544to get the proper definitions imported.  See the examples in
7545L<perlipc/"Sockets: Client/Server Communication">.
7546
7547On systems that support a close-on-exec flag on files, the flag will
7548be set for the newly opened file descriptor, as determined by the
7549value of L<C<$^F>|perlvar/$^F>.  See L<perlvar/$^F>.
7550
7551=item socketpair SOCKET1,SOCKET2,DOMAIN,TYPE,PROTOCOL
7552X<socketpair>
7553
7554=for Pod::Functions create a pair of sockets
7555
7556Creates an unnamed pair of sockets in the specified domain, of the
7557specified type.  DOMAIN, TYPE, and PROTOCOL are specified the same as
7558for the syscall of the same name.  If unimplemented, raises an exception.
7559Returns true if successful.
7560
7561On systems that support a close-on-exec flag on files, the flag will
7562be set for the newly opened file descriptors, as determined by the value
7563of L<C<$^F>|perlvar/$^F>.  See L<perlvar/$^F>.
7564
7565Some systems define L<C<pipe>|/pipe READHANDLE,WRITEHANDLE> in terms of
7566L<C<socketpair>|/socketpair SOCKET1,SOCKET2,DOMAIN,TYPE,PROTOCOL>, in
7567which a call to C<pipe($rdr, $wtr)> is essentially:
7568
7569    use Socket;
7570    socketpair(my $rdr, my $wtr, AF_UNIX, SOCK_STREAM, PF_UNSPEC);
7571    shutdown($rdr, 1);        # no more writing for reader
7572    shutdown($wtr, 0);        # no more reading for writer
7573
7574See L<perlipc> for an example of socketpair use.  Perl 5.8 and later will
7575emulate socketpair using IP sockets to localhost if your system implements
7576sockets but not socketpair.
7577
7578Portability issues: L<perlport/socketpair>.
7579
7580=item sort SUBNAME LIST
7581X<sort>
7582
7583=item sort BLOCK LIST
7584
7585=item sort LIST
7586
7587=for Pod::Functions sort a list of values
7588
7589In list context, this sorts the LIST and returns the sorted list value.
7590In scalar context, the behaviour of L<C<sort>|/sort SUBNAME LIST> is
7591undefined.
7592
7593If SUBNAME or BLOCK is omitted, L<C<sort>|/sort SUBNAME LIST>s in
7594standard string comparison
7595order.  If SUBNAME is specified, it gives the name of a subroutine
7596that returns an integer less than, equal to, or greater than C<0>,
7597depending on how the elements of the list are to be ordered.  (The
7598C<< <=> >> and C<cmp> operators are extremely useful in such routines.)
7599SUBNAME may be a scalar variable name (unsubscripted), in which case
7600the value provides the name of (or a reference to) the actual
7601subroutine to use.  In place of a SUBNAME, you can provide a BLOCK as
7602an anonymous, in-line sort subroutine.
7603
7604If the subroutine's prototype is C<($$)>, the elements to be compared are
7605passed by reference in L<C<@_>|perlvar/@_>, as for a normal subroutine.
7606This is slower than unprototyped subroutines, where the elements to be
7607compared are passed into the subroutine as the package global variables
7608C<$a> and C<$b> (see example below).
7609
7610If the subroutine is an XSUB, the elements to be compared are pushed on
7611to the stack, the way arguments are usually passed to XSUBs.  C<$a> and
7612C<$b> are not set.
7613
7614The values to be compared are always passed by reference and should not
7615be modified.
7616
7617You also cannot exit out of the sort block or subroutine using any of the
7618loop control operators described in L<perlsyn> or with
7619L<C<goto>|/goto LABEL>.
7620
7621When L<C<use locale>|locale> (but not C<use locale ':not_characters'>)
7622is in effect, C<sort LIST> sorts LIST according to the
7623current collation locale.  See L<perllocale>.
7624
7625L<C<sort>|/sort SUBNAME LIST> returns aliases into the original list,
7626much as a for loop's index variable aliases the list elements.  That is,
7627modifying an element of a list returned by L<C<sort>|/sort SUBNAME LIST>
7628(for example, in a C<foreach>, L<C<map>|/map BLOCK LIST> or
7629L<C<grep>|/grep BLOCK LIST>)
7630actually modifies the element in the original list.  This is usually
7631something to be avoided when writing clear code.
7632
7633Historically Perl has varied in whether sorting is stable by default.
7634If stability matters, it can be controlled explicitly by using the
7635L<sort> pragma.
7636
7637Examples:
7638
7639    # sort lexically
7640    my @articles = sort @files;
7641
7642    # same thing, but with explicit sort routine
7643    my @articles = sort {$a cmp $b} @files;
7644
7645    # now case-insensitively
7646    my @articles = sort {fc($a) cmp fc($b)} @files;
7647
7648    # same thing in reversed order
7649    my @articles = sort {$b cmp $a} @files;
7650
7651    # sort numerically ascending
7652    my @articles = sort {$a <=> $b} @files;
7653
7654    # sort numerically descending
7655    my @articles = sort {$b <=> $a} @files;
7656
7657    # this sorts the %age hash by value instead of key
7658    # using an in-line function
7659    my @eldest = sort { $age{$b} <=> $age{$a} } keys %age;
7660
7661    # sort using explicit subroutine name
7662    sub byage {
7663        $age{$a} <=> $age{$b};  # presuming numeric
7664    }
7665    my @sortedclass = sort byage @class;
7666
7667    sub backwards { $b cmp $a }
7668    my @harry  = qw(dog cat x Cain Abel);
7669    my @george = qw(gone chased yz Punished Axed);
7670    print sort @harry;
7671        # prints AbelCaincatdogx
7672    print sort backwards @harry;
7673        # prints xdogcatCainAbel
7674    print sort @george, 'to', @harry;
7675        # prints AbelAxedCainPunishedcatchaseddoggonetoxyz
7676
7677    # inefficiently sort by descending numeric compare using
7678    # the first integer after the first = sign, or the
7679    # whole record case-insensitively otherwise
7680
7681    my @new = sort {
7682        ($b =~ /=(\d+)/)[0] <=> ($a =~ /=(\d+)/)[0]
7683                            ||
7684                    fc($a)  cmp  fc($b)
7685    } @old;
7686
7687    # same thing, but much more efficiently;
7688    # we'll build auxiliary indices instead
7689    # for speed
7690    my (@nums, @caps);
7691    for (@old) {
7692        push @nums, ( /=(\d+)/ ? $1 : undef );
7693        push @caps, fc($_);
7694    }
7695
7696    my @new = @old[ sort {
7697                           $nums[$b] <=> $nums[$a]
7698                                    ||
7699                           $caps[$a] cmp $caps[$b]
7700                         } 0..$#old
7701                  ];
7702
7703    # same thing, but without any temps
7704    my @new = map { $_->[0] }
7705           sort { $b->[1] <=> $a->[1]
7706                           ||
7707                  $a->[2] cmp $b->[2]
7708           } map { [$_, /=(\d+)/, fc($_)] } @old;
7709
7710    # using a prototype allows you to use any comparison subroutine
7711    # as a sort subroutine (including other package's subroutines)
7712    package Other;
7713    sub backwards ($$) { $_[1] cmp $_[0]; }  # $a and $b are
7714                                             # not set here
7715    package main;
7716    my @new = sort Other::backwards @old;
7717
7718    ## using a prototype with function signature
7719    use feature 'signatures';
7720    sub function_with_signature :prototype($$) ($one, $two) {
7721        return $one <=> $two
7722    }
7723
7724    my @new = sort function_with_signature @old;
7725
7726    # guarantee stability
7727    use sort 'stable';
7728    my @new = sort { substr($a, 3, 5) cmp substr($b, 3, 5) } @old;
7729
7730Warning: syntactical care is required when sorting the list returned from
7731a function.  If you want to sort the list returned by the function call
7732C<find_records(@key)>, you can use:
7733
7734    my @contact = sort { $a cmp $b } find_records @key;
7735    my @contact = sort +find_records(@key);
7736    my @contact = sort &find_records(@key);
7737    my @contact = sort(find_records(@key));
7738
7739If instead you want to sort the array C<@key> with the comparison routine
7740C<find_records()> then you can use:
7741
7742    my @contact = sort { find_records() } @key;
7743    my @contact = sort find_records(@key);
7744    my @contact = sort(find_records @key);
7745    my @contact = sort(find_records (@key));
7746
7747C<$a> and C<$b> are set as package globals in the package the sort() is
7748called from.  That means C<$main::a> and C<$main::b> (or C<$::a> and
7749C<$::b>) in the C<main> package, C<$FooPack::a> and C<$FooPack::b> in the
7750C<FooPack> package, etc.  If the sort block is in scope of a C<my> or
7751C<state> declaration of C<$a> and/or C<$b>, you I<must> spell out the full
7752name of the variables in the sort block :
7753
7754   package main;
7755   my $a = "C"; # DANGER, Will Robinson, DANGER !!!
7756
7757   print sort { $a cmp $b }               qw(A C E G B D F H);
7758                                          # WRONG
7759   sub badlexi { $a cmp $b }
7760   print sort badlexi                     qw(A C E G B D F H);
7761                                          # WRONG
7762   # the above prints BACFEDGH or some other incorrect ordering
7763
7764   print sort { $::a cmp $::b }           qw(A C E G B D F H);
7765                                          # OK
7766   print sort { our $a cmp our $b }       qw(A C E G B D F H);
7767                                          # also OK
7768   print sort { our ($a, $b); $a cmp $b } qw(A C E G B D F H);
7769                                          # also OK
7770   sub lexi { our $a cmp our $b }
7771   print sort lexi                        qw(A C E G B D F H);
7772                                          # also OK
7773   # the above print ABCDEFGH
7774
7775With proper care you may mix package and my (or state) C<$a> and/or C<$b>:
7776
7777   my $a = {
7778      tiny   => -2,
7779      small  => -1,
7780      normal => 0,
7781      big    => 1,
7782      huge   => 2
7783   };
7784
7785   say sort { $a->{our $a} <=> $a->{our $b} }
7786       qw{ huge normal tiny small big};
7787
7788   # prints tinysmallnormalbighuge
7789
7790C<$a> and C<$b> are implicitly local to the sort() execution and regain their
7791former values upon completing the sort.
7792
7793Sort subroutines written using C<$a> and C<$b> are bound to their calling
7794package. It is possible, but of limited interest, to define them in a
7795different package, since the subroutine must still refer to the calling
7796package's C<$a> and C<$b> :
7797
7798   package Foo;
7799   sub lexi { $Bar::a cmp $Bar::b }
7800   package Bar;
7801   ... sort Foo::lexi ...
7802
7803Use the prototyped versions (see above) for a more generic alternative.
7804
7805The comparison function is required to behave.  If it returns
7806inconsistent results (sometimes saying C<$x[1]> is less than C<$x[2]> and
7807sometimes saying the opposite, for example) the results are not
7808well-defined.
7809
7810Because C<< <=> >> returns L<C<undef>|/undef EXPR> when either operand
7811is C<NaN> (not-a-number), be careful when sorting with a
7812comparison function like C<< $a <=> $b >> any lists that might contain a
7813C<NaN>.  The following example takes advantage that C<NaN != NaN> to
7814eliminate any C<NaN>s from the input list.
7815
7816    my @result = sort { $a <=> $b } grep { $_ == $_ } @input;
7817
7818In this version of F<perl>, the C<sort> function is implemented via the
7819mergesort algorithm.
7820
7821=item splice ARRAY,OFFSET,LENGTH,LIST
7822X<splice>
7823
7824=item splice ARRAY,OFFSET,LENGTH
7825
7826=item splice ARRAY,OFFSET
7827
7828=item splice ARRAY
7829
7830=for Pod::Functions add or remove elements anywhere in an array
7831
7832Removes the elements designated by OFFSET and LENGTH from an array, and
7833replaces them with the elements of LIST, if any.  In list context,
7834returns the elements removed from the array.  In scalar context,
7835returns the last element removed, or L<C<undef>|/undef EXPR> if no
7836elements are
7837removed.  The array grows or shrinks as necessary.
7838If OFFSET is negative then it starts that far from the end of the array.
7839If LENGTH is omitted, removes everything from OFFSET onward.
7840If LENGTH is negative, removes the elements from OFFSET onward
7841except for -LENGTH elements at the end of the array.
7842If both OFFSET and LENGTH are omitted, removes everything.  If OFFSET is
7843past the end of the array and a LENGTH was provided, Perl issues a warning,
7844and splices at the end of the array.
7845
7846The following equivalences hold (assuming C<< $#a >= $i >> )
7847
7848    push(@a,$x,$y)      splice(@a,@a,0,$x,$y)
7849    pop(@a)             splice(@a,-1)
7850    shift(@a)           splice(@a,0,1)
7851    unshift(@a,$x,$y)   splice(@a,0,0,$x,$y)
7852    $a[$i] = $y         splice(@a,$i,1,$y)
7853
7854L<C<splice>|/splice ARRAY,OFFSET,LENGTH,LIST> can be used, for example,
7855to implement n-ary queue processing:
7856
7857    sub nary_print {
7858      my $n = shift;
7859      while (my @next_n = splice @_, 0, $n) {
7860        say join q{ -- }, @next_n;
7861      }
7862    }
7863
7864    nary_print(3, qw(a b c d e f g h));
7865    # prints:
7866    #   a -- b -- c
7867    #   d -- e -- f
7868    #   g -- h
7869
7870Starting with Perl 5.14, an experimental feature allowed
7871L<C<splice>|/splice ARRAY,OFFSET,LENGTH,LIST> to take a
7872scalar expression. This experiment has been deemed unsuccessful, and was
7873removed as of Perl 5.24.
7874
7875=item split /PATTERN/,EXPR,LIMIT
7876X<split>
7877
7878=item split /PATTERN/,EXPR
7879
7880=item split /PATTERN/
7881
7882=item split
7883
7884=for Pod::Functions split up a string using a regexp delimiter
7885
7886Splits the string EXPR into a list of strings and returns the
7887list in list context, or the size of the list in scalar context.
7888(Prior to Perl 5.11, it also overwrote C<@_> with the list in
7889void and scalar context. If you target old perls, beware.)
7890
7891If only PATTERN is given, EXPR defaults to L<C<$_>|perlvar/$_>.
7892
7893Anything in EXPR that matches PATTERN is taken to be a separator
7894that separates the EXPR into substrings (called "I<fields>") that
7895do B<not> include the separator.  Note that a separator may be
7896longer than one character or even have no characters at all (the
7897empty string, which is a zero-width match).
7898
7899The PATTERN need not be constant; an expression may be used
7900to specify a pattern that varies at runtime.
7901
7902If PATTERN matches the empty string, the EXPR is split at the match
7903position (between characters).  As an example, the following:
7904
7905    my @x = split(/b/, "abc"); # ("a", "c")
7906
7907uses the C<b> in C<'abc'> as a separator to produce the list ("a", "c").
7908However, this:
7909
7910    my @x = split(//, "abc"); # ("a", "b", "c")
7911
7912uses empty string matches as separators; thus, the empty string
7913may be used to split EXPR into a list of its component characters.
7914
7915As a special case for L<C<split>|/split E<sol>PATTERNE<sol>,EXPR,LIMIT>,
7916the empty pattern given in
7917L<match operator|perlop/"m/PATTERN/msixpodualngc"> syntax (C<//>)
7918specifically matches the empty string, which is contrary to its usual
7919interpretation as the last successful match.
7920
7921If PATTERN is C</^/>, then it is treated as if it used the
7922L<multiline modifier|perlreref/OPERATORS> (C</^/m>), since it
7923isn't much use otherwise.
7924
7925C<E<sol>m> and any of the other pattern modifiers valid for C<qr>
7926(summarized in L<perlop/qrE<sol>STRINGE<sol>msixpodualn>) may be
7927specified explicitly.
7928
7929As another special case,
7930L<C<split>|/split E<sol>PATTERNE<sol>,EXPR,LIMIT> emulates the default
7931behavior of the
7932command line tool B<awk> when the PATTERN is either omitted or a
7933string composed of a single space character (such as S<C<' '>> or
7934S<C<"\x20">>, but not e.g. S<C</ />>).  In this case, any leading
7935whitespace in EXPR is removed before splitting occurs, and the PATTERN is
7936instead treated as if it were C</\s+/>; in particular, this means that
7937I<any> contiguous whitespace (not just a single space character) is used as
7938a separator.
7939
7940    my @x = split(" ", "  Quick brown fox\n");
7941    # ("Quick", "brown", "fox")
7942
7943    my @x = split(" ", "RED\tGREEN\tBLUE");
7944    # ("RED", "GREEN", "BLUE")
7945
7946Using split in this fashion is very similar to how
7947L<C<qwE<sol>E<sol>>|/qwE<sol>STRINGE<sol>> works.
7948
7949However, this special treatment can be avoided by specifying
7950the pattern S<C</ />> instead of the string S<C<" ">>, thereby allowing
7951only a single space character to be a separator.  In earlier Perls this
7952special case was restricted to the use of a plain S<C<" ">> as the
7953pattern argument to split; in Perl 5.18.0 and later this special case is
7954triggered by any expression which evaluates to the simple string S<C<" ">>.
7955
7956As of Perl 5.28, this special-cased whitespace splitting works as expected in
7957the scope of L<< S<C<"use feature 'unicode_strings'">>|feature/The
7958'unicode_strings' feature >>. In previous versions, and outside the scope of
7959that feature, it exhibits L<perlunicode/The "Unicode Bug">: characters that are
7960whitespace according to Unicode rules but not according to ASCII rules can be
7961treated as part of fields rather than as field separators, depending on the
7962string's internal encoding.
7963
7964If omitted, PATTERN defaults to a single space, S<C<" ">>, triggering
7965the previously described I<awk> emulation.
7966
7967If LIMIT is specified and positive, it represents the maximum number
7968of fields into which the EXPR may be split; in other words, LIMIT is
7969one greater than the maximum number of times EXPR may be split.  Thus,
7970the LIMIT value C<1> means that EXPR may be split a maximum of zero
7971times, producing a maximum of one field (namely, the entire value of
7972EXPR).  For instance:
7973
7974    my @x = split(//, "abc", 1); # ("abc")
7975    my @x = split(//, "abc", 2); # ("a", "bc")
7976    my @x = split(//, "abc", 3); # ("a", "b", "c")
7977    my @x = split(//, "abc", 4); # ("a", "b", "c")
7978
7979If LIMIT is negative, it is treated as if it were instead arbitrarily
7980large; as many fields as possible are produced.
7981
7982If LIMIT is omitted (or, equivalently, zero), then it is usually
7983treated as if it were instead negative but with the exception that
7984trailing empty fields are stripped (empty leading fields are always
7985preserved); if all fields are empty, then all fields are considered to
7986be trailing (and are thus stripped in this case).  Thus, the following:
7987
7988    my @x = split(/,/, "a,b,c,,,"); # ("a", "b", "c")
7989
7990produces only a three element list.
7991
7992    my @x = split(/,/, "a,b,c,,,", -1); # ("a", "b", "c", "", "", "")
7993
7994produces a six element list.
7995
7996In time-critical applications, it is worthwhile to avoid splitting
7997into more fields than necessary.  Thus, when assigning to a list,
7998if LIMIT is omitted (or zero), then LIMIT is treated as though it
7999were one larger than the number of variables in the list; for the
8000following, LIMIT is implicitly 3:
8001
8002    my ($login, $passwd) = split(/:/);
8003
8004Note that splitting an EXPR that evaluates to the empty string always
8005produces zero fields, regardless of the LIMIT specified.
8006
8007An empty leading field is produced when there is a positive-width
8008match at the beginning of EXPR.  For instance:
8009
8010    my @x = split(/ /, " abc"); # ("", "abc")
8011
8012splits into two elements.  However, a zero-width match at the
8013beginning of EXPR never produces an empty field, so that:
8014
8015    my @x = split(//, " abc"); # (" ", "a", "b", "c")
8016
8017splits into four elements instead of five.
8018
8019An empty trailing field, on the other hand, is produced when there is a
8020match at the end of EXPR, regardless of the length of the match
8021(of course, unless a non-zero LIMIT is given explicitly, such fields are
8022removed, as in the last example).  Thus:
8023
8024    my @x = split(//, " abc", -1); # (" ", "a", "b", "c", "")
8025
8026If the PATTERN contains
8027L<capturing groups|perlretut/Grouping things and hierarchical matching>,
8028then for each separator, an additional field is produced for each substring
8029captured by a group (in the order in which the groups are specified,
8030as per L<backreferences|perlretut/Backreferences>); if any group does not
8031match, then it captures the L<C<undef>|/undef EXPR> value instead of a
8032substring.  Also,
8033note that any such additional field is produced whenever there is a
8034separator (that is, whenever a split occurs), and such an additional field
8035does B<not> count towards the LIMIT.  Consider the following expressions
8036evaluated in list context (each returned list is provided in the associated
8037comment):
8038
8039    my @x = split(/-|,/    , "1-10,20", 3);
8040    # ("1", "10", "20")
8041
8042    my @x = split(/(-|,)/  , "1-10,20", 3);
8043    # ("1", "-", "10", ",", "20")
8044
8045    my @x = split(/-|(,)/  , "1-10,20", 3);
8046    # ("1", undef, "10", ",", "20")
8047
8048    my @x = split(/(-)|,/  , "1-10,20", 3);
8049    # ("1", "-", "10", undef, "20")
8050
8051    my @x = split(/(-)|(,)/, "1-10,20", 3);
8052    # ("1", "-", undef, "10", undef, ",", "20")
8053
8054=item sprintf FORMAT, LIST
8055X<sprintf>
8056
8057=for Pod::Functions formatted print into a string
8058
8059Returns a string formatted by the usual
8060L<C<printf>|/printf FILEHANDLE FORMAT, LIST> conventions of the C
8061library function L<C<sprintf>|/sprintf FORMAT, LIST>.  See below for
8062more details and see L<sprintf(3)> or L<printf(3)> on your system for an
8063explanation of the general principles.
8064
8065For example:
8066
8067        # Format number with up to 8 leading zeroes
8068        my $result = sprintf("%08d", $number);
8069
8070        # Round number to 3 digits after decimal point
8071        my $rounded = sprintf("%.3f", $number);
8072
8073Perl does its own L<C<sprintf>|/sprintf FORMAT, LIST> formatting: it
8074emulates the C
8075function L<sprintf(3)>, but doesn't use it except for floating-point
8076numbers, and even then only standard modifiers are allowed.
8077Non-standard extensions in your local L<sprintf(3)> are
8078therefore unavailable from Perl.
8079
8080Unlike L<C<printf>|/printf FILEHANDLE FORMAT, LIST>,
8081L<C<sprintf>|/sprintf FORMAT, LIST> does not do what you probably mean
8082when you pass it an array as your first argument.
8083The array is given scalar context,
8084and instead of using the 0th element of the array as the format, Perl will
8085use the count of elements in the array as the format, which is almost never
8086useful.
8087
8088Perl's L<C<sprintf>|/sprintf FORMAT, LIST> permits the following
8089universally-known conversions:
8090
8091   %%    a percent sign
8092   %c    a character with the given number
8093   %s    a string
8094   %d    a signed integer, in decimal
8095   %u    an unsigned integer, in decimal
8096   %o    an unsigned integer, in octal
8097   %x    an unsigned integer, in hexadecimal
8098   %e    a floating-point number, in scientific notation
8099   %f    a floating-point number, in fixed decimal notation
8100   %g    a floating-point number, in %e or %f notation
8101
8102In addition, Perl permits the following widely-supported conversions:
8103
8104   %X    like %x, but using upper-case letters
8105   %E    like %e, but using an upper-case "E"
8106   %G    like %g, but with an upper-case "E" (if applicable)
8107   %b    an unsigned integer, in binary
8108   %B    like %b, but using an upper-case "B" with the # flag
8109   %p    a pointer (outputs the Perl value's address in hexadecimal)
8110   %n    special: *stores* the number of characters output so far
8111         into the next argument in the parameter list
8112   %a    hexadecimal floating point
8113   %A    like %a, but using upper-case letters
8114
8115Finally, for backward (and we do mean "backward") compatibility, Perl
8116permits these unnecessary but widely-supported conversions:
8117
8118   %i    a synonym for %d
8119   %D    a synonym for %ld
8120   %U    a synonym for %lu
8121   %O    a synonym for %lo
8122   %F    a synonym for %f
8123
8124Note that the number of exponent digits in the scientific notation produced
8125by C<%e>, C<%E>, C<%g> and C<%G> for numbers with the modulus of the
8126exponent less than 100 is system-dependent: it may be three or less
8127(zero-padded as necessary).  In other words, 1.23 times ten to the
812899th may be either "1.23e99" or "1.23e099".  Similarly for C<%a> and C<%A>:
8129the exponent or the hexadecimal digits may float: especially the
8130"long doubles" Perl configuration option may cause surprises.
8131
8132Between the C<%> and the format letter, you may specify several
8133additional attributes controlling the interpretation of the format.
8134In order, these are:
8135
8136=over 4
8137
8138=item format parameter index
8139
8140An explicit format parameter index, such as C<2$>.  By default sprintf
8141will format the next unused argument in the list, but this allows you
8142to take the arguments out of order:
8143
8144  printf '%2$d %1$d', 12, 34;      # prints "34 12"
8145  printf '%3$d %d %1$d', 1, 2, 3;  # prints "3 1 1"
8146
8147=item flags
8148
8149one or more of:
8150
8151   space   prefix non-negative number with a space
8152   +       prefix non-negative number with a plus sign
8153   -       left-justify within the field
8154   0       use zeros, not spaces, to right-justify
8155   #       ensure the leading "0" for any octal,
8156           prefix non-zero hexadecimal with "0x" or "0X",
8157           prefix non-zero binary with "0b" or "0B"
8158
8159For example:
8160
8161  printf '<% d>',  12;   # prints "< 12>"
8162  printf '<% d>',   0;   # prints "< 0>"
8163  printf '<% d>', -12;   # prints "<-12>"
8164  printf '<%+d>',  12;   # prints "<+12>"
8165  printf '<%+d>',   0;   # prints "<+0>"
8166  printf '<%+d>', -12;   # prints "<-12>"
8167  printf '<%6s>',  12;   # prints "<    12>"
8168  printf '<%-6s>', 12;   # prints "<12    >"
8169  printf '<%06s>', 12;   # prints "<000012>"
8170  printf '<%#o>',  12;   # prints "<014>"
8171  printf '<%#x>',  12;   # prints "<0xc>"
8172  printf '<%#X>',  12;   # prints "<0XC>"
8173  printf '<%#b>',  12;   # prints "<0b1100>"
8174  printf '<%#B>',  12;   # prints "<0B1100>"
8175
8176When a space and a plus sign are given as the flags at once,
8177the space is ignored.
8178
8179  printf '<%+ d>', 12;   # prints "<+12>"
8180  printf '<% +d>', 12;   # prints "<+12>"
8181
8182When the # flag and a precision are given in the %o conversion,
8183the precision is incremented if it's necessary for the leading "0".
8184
8185  printf '<%#.5o>', 012;      # prints "<00012>"
8186  printf '<%#.5o>', 012345;   # prints "<012345>"
8187  printf '<%#.0o>', 0;        # prints "<0>"
8188
8189=item vector flag
8190
8191This flag tells Perl to interpret the supplied string as a vector of
8192integers, one for each character in the string.  Perl applies the format to
8193each integer in turn, then joins the resulting strings with a separator (a
8194dot C<.> by default).  This can be useful for displaying ordinal values of
8195characters in arbitrary strings:
8196
8197  printf "%vd", "AB\x{100}";           # prints "65.66.256"
8198  printf "version is v%vd\n", $^V;     # Perl's version
8199
8200Put an asterisk C<*> before the C<v> to override the string to
8201use to separate the numbers:
8202
8203  printf "address is %*vX\n", ":", $addr;   # IPv6 address
8204  printf "bits are %0*v8b\n", " ", $bits;   # random bitstring
8205
8206You can also explicitly specify the argument number to use for
8207the join string using something like C<*2$v>; for example:
8208
8209  printf '%*4$vX %*4$vX %*4$vX',       # 3 IPv6 addresses
8210          @addr[1..3], ":";
8211
8212=item (minimum) width
8213
8214Arguments are usually formatted to be only as wide as required to
8215display the given value.  You can override the width by putting
8216a number here, or get the width from the next argument (with C<*>)
8217or from a specified argument (e.g., with C<*2$>):
8218
8219 printf "<%s>", "a";       # prints "<a>"
8220 printf "<%6s>", "a";      # prints "<     a>"
8221 printf "<%*s>", 6, "a";   # prints "<     a>"
8222 printf '<%*2$s>', "a", 6; # prints "<     a>"
8223 printf "<%2s>", "long";   # prints "<long>" (does not truncate)
8224
8225If a field width obtained through C<*> is negative, it has the same
8226effect as the C<-> flag: left-justification.
8227
8228=item precision, or maximum width
8229X<precision>
8230
8231You can specify a precision (for numeric conversions) or a maximum
8232width (for string conversions) by specifying a C<.> followed by a number.
8233For floating-point formats except C<g> and C<G>, this specifies
8234how many places right of the decimal point to show (the default being 6).
8235For example:
8236
8237  # these examples are subject to system-specific variation
8238  printf '<%f>', 1;    # prints "<1.000000>"
8239  printf '<%.1f>', 1;  # prints "<1.0>"
8240  printf '<%.0f>', 1;  # prints "<1>"
8241  printf '<%e>', 10;   # prints "<1.000000e+01>"
8242  printf '<%.1e>', 10; # prints "<1.0e+01>"
8243
8244For "g" and "G", this specifies the maximum number of significant digits to
8245show; for example:
8246
8247  # These examples are subject to system-specific variation.
8248  printf '<%g>', 1;        # prints "<1>"
8249  printf '<%.10g>', 1;     # prints "<1>"
8250  printf '<%g>', 100;      # prints "<100>"
8251  printf '<%.1g>', 100;    # prints "<1e+02>"
8252  printf '<%.2g>', 100.01; # prints "<1e+02>"
8253  printf '<%.5g>', 100.01; # prints "<100.01>"
8254  printf '<%.4g>', 100.01; # prints "<100>"
8255  printf '<%.1g>', 0.0111; # prints "<0.01>"
8256  printf '<%.2g>', 0.0111; # prints "<0.011>"
8257  printf '<%.3g>', 0.0111; # prints "<0.0111>"
8258
8259For integer conversions, specifying a precision implies that the
8260output of the number itself should be zero-padded to this width,
8261where the 0 flag is ignored:
8262
8263  printf '<%.6d>', 1;      # prints "<000001>"
8264  printf '<%+.6d>', 1;     # prints "<+000001>"
8265  printf '<%-10.6d>', 1;   # prints "<000001    >"
8266  printf '<%10.6d>', 1;    # prints "<    000001>"
8267  printf '<%010.6d>', 1;   # prints "<    000001>"
8268  printf '<%+10.6d>', 1;   # prints "<   +000001>"
8269
8270  printf '<%.6x>', 1;      # prints "<000001>"
8271  printf '<%#.6x>', 1;     # prints "<0x000001>"
8272  printf '<%-10.6x>', 1;   # prints "<000001    >"
8273  printf '<%10.6x>', 1;    # prints "<    000001>"
8274  printf '<%010.6x>', 1;   # prints "<    000001>"
8275  printf '<%#10.6x>', 1;   # prints "<  0x000001>"
8276
8277For string conversions, specifying a precision truncates the string
8278to fit the specified width:
8279
8280  printf '<%.5s>', "truncated";   # prints "<trunc>"
8281  printf '<%10.5s>', "truncated"; # prints "<     trunc>"
8282
8283You can also get the precision from the next argument using C<.*>, or from a
8284specified argument (e.g., with C<.*2$>):
8285
8286  printf '<%.6x>', 1;       # prints "<000001>"
8287  printf '<%.*x>', 6, 1;    # prints "<000001>"
8288
8289  printf '<%.*2$x>', 1, 6;  # prints "<000001>"
8290
8291  printf '<%6.*2$x>', 1, 4; # prints "<  0001>"
8292
8293If a precision obtained through C<*> is negative, it counts
8294as having no precision at all.
8295
8296  printf '<%.*s>',  7, "string";   # prints "<string>"
8297  printf '<%.*s>',  3, "string";   # prints "<str>"
8298  printf '<%.*s>',  0, "string";   # prints "<>"
8299  printf '<%.*s>', -1, "string";   # prints "<string>"
8300
8301  printf '<%.*d>',  1, 0;   # prints "<0>"
8302  printf '<%.*d>',  0, 0;   # prints "<>"
8303  printf '<%.*d>', -1, 0;   # prints "<0>"
8304
8305=item size
8306
8307For numeric conversions, you can specify the size to interpret the
8308number as using C<l>, C<h>, C<V>, C<q>, C<L>, or C<ll>.  For integer
8309conversions (C<d u o x X b i D U O>), numbers are usually assumed to be
8310whatever the default integer size is on your platform (usually 32 or 64
8311bits), but you can override this to use instead one of the standard C types,
8312as supported by the compiler used to build Perl:
8313
8314   hh          interpret integer as C type "char" or "unsigned
8315               char" on Perl 5.14 or later
8316   h           interpret integer as C type "short" or
8317               "unsigned short"
8318   j           interpret integer as C type "intmax_t" on Perl
8319               5.14 or later; and prior to Perl 5.30, only with
8320               a C99 compiler (unportable)
8321   l           interpret integer as C type "long" or
8322               "unsigned long"
8323   q, L, or ll interpret integer as C type "long long",
8324               "unsigned long long", or "quad" (typically
8325               64-bit integers)
8326   t           interpret integer as C type "ptrdiff_t" on Perl
8327               5.14 or later
8328   z           interpret integer as C types "size_t" or
8329               "ssize_t" on Perl 5.14 or later
8330
8331Note that, in general, using the C<l> modifier (for example, when writing
8332C<"%ld"> or C<"%lu"> instead of C<"%d"> and C<"%u">) is unnecessary
8333when used from Perl code.  Moreover, it may be harmful, for example on
8334Windows 64-bit where a long is 32-bits.
8335
8336As of 5.14, none of these raises an exception if they are not supported on
8337your platform.  However, if warnings are enabled, a warning of the
8338L<C<printf>|warnings> warning class is issued on an unsupported
8339conversion flag.  Should you instead prefer an exception, do this:
8340
8341    use warnings FATAL => "printf";
8342
8343If you would like to know about a version dependency before you
8344start running the program, put something like this at its top:
8345
8346    use v5.14;  # for hh/j/t/z/ printf modifiers
8347
8348You can find out whether your Perl supports quads via L<Config>:
8349
8350    use Config;
8351    if ($Config{use64bitint} eq "define"
8352        || $Config{longsize} >= 8) {
8353        print "Nice quads!\n";
8354    }
8355
8356For floating-point conversions (C<e f g E F G>), numbers are usually assumed
8357to be the default floating-point size on your platform (double or long double),
8358but you can force "long double" with C<q>, C<L>, or C<ll> if your
8359platform supports them.  You can find out whether your Perl supports long
8360doubles via L<Config>:
8361
8362    use Config;
8363    print "long doubles\n" if $Config{d_longdbl} eq "define";
8364
8365You can find out whether Perl considers "long double" to be the default
8366floating-point size to use on your platform via L<Config>:
8367
8368    use Config;
8369    if ($Config{uselongdouble} eq "define") {
8370        print "long doubles by default\n";
8371    }
8372
8373It can also be that long doubles and doubles are the same thing:
8374
8375        use Config;
8376        ($Config{doublesize} == $Config{longdblsize}) &&
8377                print "doubles are long doubles\n";
8378
8379The size specifier C<V> has no effect for Perl code, but is supported for
8380compatibility with XS code.  It means "use the standard size for a Perl
8381integer or floating-point number", which is the default.
8382
8383=item order of arguments
8384
8385Normally, L<C<sprintf>|/sprintf FORMAT, LIST> takes the next unused
8386argument as the value to
8387format for each format specification.  If the format specification
8388uses C<*> to require additional arguments, these are consumed from
8389the argument list in the order they appear in the format
8390specification I<before> the value to format.  Where an argument is
8391specified by an explicit index, this does not affect the normal
8392order for the arguments, even when the explicitly specified index
8393would have been the next argument.
8394
8395So:
8396
8397    printf "<%*.*s>", $a, $b, $c;
8398
8399uses C<$a> for the width, C<$b> for the precision, and C<$c>
8400as the value to format; while:
8401
8402  printf '<%*1$.*s>', $a, $b;
8403
8404would use C<$a> for the width and precision, and C<$b> as the
8405value to format.
8406
8407Here are some more examples; be aware that when using an explicit
8408index, the C<$> may need escaping:
8409
8410 printf "%2\$d %d\n",      12, 34;     # will print "34 12\n"
8411 printf "%2\$d %d %d\n",   12, 34;     # will print "34 12 34\n"
8412 printf "%3\$d %d %d\n",   12, 34, 56; # will print "56 12 34\n"
8413 printf "%2\$*3\$d %d\n",  12, 34,  3; # will print " 34 12\n"
8414 printf "%*1\$.*f\n",       4,  5, 10; # will print "5.0000\n"
8415
8416=back
8417
8418If L<C<use locale>|locale> (including C<use locale ':not_characters'>)
8419is in effect and L<C<POSIX::setlocale>|POSIX/C<setlocale>> has been
8420called,
8421the character used for the decimal separator in formatted floating-point
8422numbers is affected by the C<LC_NUMERIC> locale.  See L<perllocale>
8423and L<POSIX>.
8424
8425=item sqrt EXPR
8426X<sqrt> X<root> X<square root>
8427
8428=item sqrt
8429
8430=for Pod::Functions square root function
8431
8432Return the positive square root of EXPR.  If EXPR is omitted, uses
8433L<C<$_>|perlvar/$_>.  Works only for non-negative operands unless you've
8434loaded the L<C<Math::Complex>|Math::Complex> module.
8435
8436    use Math::Complex;
8437    print sqrt(-4);    # prints 2i
8438
8439=item srand EXPR
8440X<srand> X<seed> X<randseed>
8441
8442=item srand
8443
8444=for Pod::Functions seed the random number generator
8445
8446Sets and returns the random number seed for the L<C<rand>|/rand EXPR>
8447operator.
8448
8449The point of the function is to "seed" the L<C<rand>|/rand EXPR>
8450function so that L<C<rand>|/rand EXPR> can produce a different sequence
8451each time you run your program.  When called with a parameter,
8452L<C<srand>|/srand EXPR> uses that for the seed; otherwise it
8453(semi-)randomly chooses a seed.  In either case, starting with Perl 5.14,
8454it returns the seed.  To signal that your code will work I<only> on Perls
8455of a recent vintage:
8456
8457    use v5.14;	# so srand returns the seed
8458
8459If L<C<srand>|/srand EXPR> is not called explicitly, it is called
8460implicitly without a parameter at the first use of the
8461L<C<rand>|/rand EXPR> operator.  However, there are a few situations
8462where programs are likely to want to call L<C<srand>|/srand EXPR>.  One
8463is for generating predictable results, generally for testing or
8464debugging.  There, you use C<srand($seed)>, with the same C<$seed> each
8465time.  Another case is that you may want to call L<C<srand>|/srand EXPR>
8466after a L<C<fork>|/fork> to avoid child processes sharing the same seed
8467value as the parent (and consequently each other).
8468
8469Do B<not> call C<srand()> (i.e., without an argument) more than once per
8470process.  The internal state of the random number generator should
8471contain more entropy than can be provided by any seed, so calling
8472L<C<srand>|/srand EXPR> again actually I<loses> randomness.
8473
8474Most implementations of L<C<srand>|/srand EXPR> take an integer and will
8475silently
8476truncate decimal numbers.  This means C<srand(42)> will usually
8477produce the same results as C<srand(42.1)>.  To be safe, always pass
8478L<C<srand>|/srand EXPR> an integer.
8479
8480A typical use of the returned seed is for a test program which has too many
8481combinations to test comprehensively in the time available to it each run.  It
8482can test a random subset each time, and should there be a failure, log the seed
8483used for that run so that it can later be used to reproduce the same results.
8484
8485B<L<C<rand>|/rand EXPR> is not cryptographically secure.  You should not rely
8486on it in security-sensitive situations.>  As of this writing, a
8487number of third-party CPAN modules offer random number generators
8488intended by their authors to be cryptographically secure,
8489including: L<Data::Entropy>, L<Crypt::Random>, L<Math::Random::Secure>,
8490and L<Math::TrulyRandom>.
8491
8492=item stat FILEHANDLE
8493X<stat> X<file, status> X<ctime>
8494
8495=item stat EXPR
8496
8497=item stat DIRHANDLE
8498
8499=item stat
8500
8501=for Pod::Functions get a file's status information
8502
8503Returns a 13-element list giving the status info for a file, either
8504the file opened via FILEHANDLE or DIRHANDLE, or named by EXPR.  If EXPR is
8505omitted, it stats L<C<$_>|perlvar/$_> (not C<_>!).  Returns the empty
8506list if L<C<stat>|/stat FILEHANDLE> fails.  Typically
8507used as follows:
8508
8509    my ($dev,$ino,$mode,$nlink,$uid,$gid,$rdev,$size,
8510        $atime,$mtime,$ctime,$blksize,$blocks)
8511           = stat($filename);
8512
8513Not all fields are supported on all filesystem types.  Here are the
8514meanings of the fields:
8515
8516  0 dev      device number of filesystem
8517  1 ino      inode number
8518  2 mode     file mode  (type and permissions)
8519  3 nlink    number of (hard) links to the file
8520  4 uid      numeric user ID of file's owner
8521  5 gid      numeric group ID of file's owner
8522  6 rdev     the device identifier (special files only)
8523  7 size     total size of file, in bytes
8524  8 atime    last access time in seconds since the epoch
8525  9 mtime    last modify time in seconds since the epoch
8526 10 ctime    inode change time in seconds since the epoch (*)
8527 11 blksize  preferred I/O size in bytes for interacting with the
8528             file (may vary from file to file)
8529 12 blocks   actual number of system-specific blocks allocated
8530             on disk (often, but not always, 512 bytes each)
8531
8532(The epoch was at 00:00 January 1, 1970 GMT.)
8533
8534(*) Not all fields are supported on all filesystem types.  Notably, the
8535ctime field is non-portable.  In particular, you cannot expect it to be a
8536"creation time"; see L<perlport/"Files and Filesystems"> for details.
8537
8538If L<C<stat>|/stat FILEHANDLE> is passed the special filehandle
8539consisting of an underline, no stat is done, but the current contents of
8540the stat structure from the last L<C<stat>|/stat FILEHANDLE>,
8541L<C<lstat>|/lstat FILEHANDLE>, or filetest are returned.  Example:
8542
8543    if (-x $file && (($d) = stat(_)) && $d < 0) {
8544        print "$file is executable NFS file\n";
8545    }
8546
8547(This works on machines only for which the device number is negative
8548under NFS.)
8549
8550On some platforms inode numbers are of a type larger than perl knows how
8551to handle as integer numerical values.  If necessary, an inode number will
8552be returned as a decimal string in order to preserve the entire value.
8553If used in a numeric context, this will be converted to a floating-point
8554numerical value, with rounding, a fate that is best avoided.  Therefore,
8555you should prefer to compare inode numbers using C<eq> rather than C<==>.
8556C<eq> will work fine on inode numbers that are represented numerically,
8557as well as those represented as strings.
8558
8559Because the mode contains both the file type and its permissions, you
8560should mask off the file type portion and (s)printf using a C<"%o">
8561if you want to see the real permissions.
8562
8563    my $mode = (stat($filename))[2];
8564    printf "Permissions are %04o\n", $mode & 07777;
8565
8566In scalar context, L<C<stat>|/stat FILEHANDLE> returns a boolean value
8567indicating success
8568or failure, and, if successful, sets the information associated with
8569the special filehandle C<_>.
8570
8571The L<File::stat> module provides a convenient, by-name access mechanism:
8572
8573    use File::stat;
8574    my $sb = stat($filename);
8575    printf "File is %s, size is %s, perm %04o, mtime %s\n",
8576           $filename, $sb->size, $sb->mode & 07777,
8577           scalar localtime $sb->mtime;
8578
8579You can import symbolic mode constants (C<S_IF*>) and functions
8580(C<S_IS*>) from the L<Fcntl> module:
8581
8582    use Fcntl ':mode';
8583
8584    my $mode = (stat($filename))[2];
8585
8586    my $user_rwx      = ($mode & S_IRWXU) >> 6;
8587    my $group_read    = ($mode & S_IRGRP) >> 3;
8588    my $other_execute =  $mode & S_IXOTH;
8589
8590    printf "Permissions are %04o\n", S_IMODE($mode), "\n";
8591
8592    my $is_setuid     =  $mode & S_ISUID;
8593    my $is_directory  =  S_ISDIR($mode);
8594
8595You could write the last two using the C<-u> and C<-d> operators.
8596Commonly available C<S_IF*> constants are:
8597
8598    # Permissions: read, write, execute, for user, group, others.
8599
8600    S_IRWXU S_IRUSR S_IWUSR S_IXUSR
8601    S_IRWXG S_IRGRP S_IWGRP S_IXGRP
8602    S_IRWXO S_IROTH S_IWOTH S_IXOTH
8603
8604    # Setuid/Setgid/Stickiness/SaveText.
8605    # Note that the exact meaning of these is system-dependent.
8606
8607    S_ISUID S_ISGID S_ISVTX S_ISTXT
8608
8609    # File types.  Not all are necessarily available on
8610    # your system.
8611
8612    S_IFREG S_IFDIR S_IFLNK S_IFBLK S_IFCHR
8613    S_IFIFO S_IFSOCK S_IFWHT S_ENFMT
8614
8615    # The following are compatibility aliases for S_IRUSR,
8616    # S_IWUSR, and S_IXUSR.
8617
8618    S_IREAD S_IWRITE S_IEXEC
8619
8620and the C<S_IF*> functions are
8621
8622    S_IMODE($mode)    the part of $mode containing the permission
8623                      bits and the setuid/setgid/sticky bits
8624
8625    S_IFMT($mode)     the part of $mode containing the file type
8626                      which can be bit-anded with (for example)
8627                      S_IFREG or with the following functions
8628
8629    # The operators -f, -d, -l, -b, -c, -p, and -S.
8630
8631    S_ISREG($mode) S_ISDIR($mode) S_ISLNK($mode)
8632    S_ISBLK($mode) S_ISCHR($mode) S_ISFIFO($mode) S_ISSOCK($mode)
8633
8634    # No direct -X operator counterpart, but for the first one
8635    # the -g operator is often equivalent.  The ENFMT stands for
8636    # record flocking enforcement, a platform-dependent feature.
8637
8638    S_ISENFMT($mode) S_ISWHT($mode)
8639
8640See your native L<chmod(2)> and L<stat(2)> documentation for more details
8641about the C<S_*> constants.  To get status info for a symbolic link
8642instead of the target file behind the link, use the
8643L<C<lstat>|/lstat FILEHANDLE> function.
8644
8645Portability issues: L<perlport/stat>.
8646
8647=item state VARLIST
8648X<state>
8649
8650=item state TYPE VARLIST
8651
8652=item state VARLIST : ATTRS
8653
8654=item state TYPE VARLIST : ATTRS
8655
8656=for Pod::Functions +state declare and assign a persistent lexical variable
8657
8658L<C<state>|/state VARLIST> declares a lexically scoped variable, just
8659like L<C<my>|/my VARLIST>.
8660However, those variables will never be reinitialized, contrary to
8661lexical variables that are reinitialized each time their enclosing block
8662is entered.
8663See L<perlsub/"Persistent Private Variables"> for details.
8664
8665If more than one variable is listed, the list must be placed in
8666parentheses.  With a parenthesised list, L<C<undef>|/undef EXPR> can be
8667used as a
8668dummy placeholder.  However, since initialization of state variables in
8669such lists is currently not possible this would serve no purpose.
8670
8671Redeclaring a variable in the same scope or statement will "shadow" the
8672previous declaration, creating a new instance and preventing access to
8673the previous one. This is usually undesired and, if warnings are enabled,
8674will result in a warning in the C<shadow> category.
8675
8676L<C<state>|/state VARLIST> is available only if the
8677L<C<"state"> feature|feature/The 'state' feature> is enabled or if it is
8678prefixed with C<CORE::>.  The
8679L<C<"state"> feature|feature/The 'state' feature> is enabled
8680automatically with a C<use v5.10> (or higher) declaration in the current
8681scope.
8682
8683
8684=item study SCALAR
8685X<study>
8686
8687=item study
8688
8689=for Pod::Functions no-op, formerly optimized input data for repeated searches
8690
8691At this time, C<study> does nothing. This may change in the future.
8692
8693Prior to Perl version 5.16, it would create an inverted index of all characters
8694that occurred in the given SCALAR (or L<C<$_>|perlvar/$_> if unspecified). When
8695matching a pattern, the rarest character from the pattern would be looked up in
8696this index. Rarity was based on some static frequency tables constructed from
8697some C programs and English text.
8698
8699
8700=item sub NAME BLOCK
8701X<sub>
8702
8703=item sub NAME (PROTO) BLOCK
8704
8705=item sub NAME : ATTRS BLOCK
8706
8707=item sub NAME (PROTO) : ATTRS BLOCK
8708
8709=for Pod::Functions declare a subroutine, possibly anonymously
8710
8711This is subroutine definition, not a real function I<per se>.  Without a
8712BLOCK it's just a forward declaration.  Without a NAME, it's an anonymous
8713function declaration, so does return a value: the CODE ref of the closure
8714just created.
8715
8716See L<perlsub> and L<perlref> for details about subroutines and
8717references; see L<attributes> and L<Attribute::Handlers> for more
8718information about attributes.
8719
8720=item __SUB__
8721X<__SUB__>
8722
8723=for Pod::Functions +current_sub the current subroutine, or C<undef> if not in a subroutine
8724
8725A special token that returns a reference to the current subroutine, or
8726L<C<undef>|/undef EXPR> outside of a subroutine.
8727
8728The behaviour of L<C<__SUB__>|/__SUB__> within a regex code block (such
8729as C</(?{...})/>) is subject to change.
8730
8731This token is only available under C<use v5.16> or the
8732L<C<"current_sub"> feature|feature/The 'current_sub' feature>.
8733See L<feature>.
8734
8735=item substr EXPR,OFFSET,LENGTH,REPLACEMENT
8736X<substr> X<substring> X<mid> X<left> X<right>
8737
8738=item substr EXPR,OFFSET,LENGTH
8739
8740=item substr EXPR,OFFSET
8741
8742=for Pod::Functions get or alter a portion of a string
8743
8744Extracts a substring out of EXPR and returns it.  First character is at
8745offset zero.  If OFFSET is negative, starts
8746that far back from the end of the string.  If LENGTH is omitted, returns
8747everything through the end of the string.  If LENGTH is negative, leaves that
8748many characters off the end of the string.
8749
8750    my $s = "The black cat climbed the green tree";
8751    my $color  = substr $s, 4, 5;      # black
8752    my $middle = substr $s, 4, -11;    # black cat climbed the
8753    my $end    = substr $s, 14;        # climbed the green tree
8754    my $tail   = substr $s, -4;        # tree
8755    my $z      = substr $s, -4, 2;     # tr
8756
8757You can use the L<C<substr>|/substr EXPR,OFFSET,LENGTH,REPLACEMENT>
8758function as an lvalue, in which case EXPR
8759must itself be an lvalue.  If you assign something shorter than LENGTH,
8760the string will shrink, and if you assign something longer than LENGTH,
8761the string will grow to accommodate it.  To keep the string the same
8762length, you may need to pad or chop your value using
8763L<C<sprintf>|/sprintf FORMAT, LIST>.
8764
8765If OFFSET and LENGTH specify a substring that is partly outside the
8766string, only the part within the string is returned.  If the substring
8767is beyond either end of the string,
8768L<C<substr>|/substr EXPR,OFFSET,LENGTH,REPLACEMENT> returns the undefined
8769value and produces a warning.  When used as an lvalue, specifying a
8770substring that is entirely outside the string raises an exception.
8771Here's an example showing the behavior for boundary cases:
8772
8773    my $name = 'fred';
8774    substr($name, 4) = 'dy';         # $name is now 'freddy'
8775    my $null = substr $name, 6, 2;   # returns "" (no warning)
8776    my $oops = substr $name, 7;      # returns undef, with warning
8777    substr($name, 7) = 'gap';        # raises an exception
8778
8779An alternative to using
8780L<C<substr>|/substr EXPR,OFFSET,LENGTH,REPLACEMENT> as an lvalue is to
8781specify the
8782replacement string as the 4th argument.  This allows you to replace
8783parts of the EXPR and return what was there before in one operation,
8784just as you can with
8785L<C<splice>|/splice ARRAY,OFFSET,LENGTH,LIST>.
8786
8787    my $s = "The black cat climbed the green tree";
8788    my $z = substr $s, 14, 7, "jumped from";    # climbed
8789    # $s is now "The black cat jumped from the green tree"
8790
8791Note that the lvalue returned by the three-argument version of
8792L<C<substr>|/substr EXPR,OFFSET,LENGTH,REPLACEMENT> acts as
8793a 'magic bullet'; each time it is assigned to, it remembers which part
8794of the original string is being modified; for example:
8795
8796    my $x = '1234';
8797    for (substr($x,1,2)) {
8798        $_ = 'a';   print $x,"\n";    # prints 1a4
8799        $_ = 'xyz'; print $x,"\n";    # prints 1xyz4
8800        $x = '56789';
8801        $_ = 'pq';  print $x,"\n";    # prints 5pq9
8802    }
8803
8804With negative offsets, it remembers its position from the end of the string
8805when the target string is modified:
8806
8807    my $x = '1234';
8808    for (substr($x, -3, 2)) {
8809        $_ = 'a';   print $x,"\n";    # prints 1a4, as above
8810        $x = 'abcdefg';
8811        print $_,"\n";                # prints f
8812    }
8813
8814Prior to Perl version 5.10, the result of using an lvalue multiple times was
8815unspecified.  Prior to 5.16, the result with negative offsets was
8816unspecified.
8817
8818=item symlink OLDFILE,NEWFILE
8819X<symlink> X<link> X<symbolic link> X<link, symbolic>
8820
8821=for Pod::Functions create a symbolic link to a file
8822
8823Creates a new filename symbolically linked to the old filename.
8824Returns C<1> for success, C<0> otherwise.  On systems that don't support
8825symbolic links, raises an exception.  To check for that,
8826use eval:
8827
8828    my $symlink_exists = eval { symlink("",""); 1 };
8829
8830Portability issues: L<perlport/symlink>.
8831
8832=item syscall NUMBER, LIST
8833X<syscall> X<system call>
8834
8835=for Pod::Functions execute an arbitrary system call
8836
8837Calls the system call specified as the first element of the list,
8838passing the remaining elements as arguments to the system call.  If
8839unimplemented, raises an exception.  The arguments are interpreted
8840as follows: if a given argument is numeric, the argument is passed as
8841an int.  If not, the pointer to the string value is passed.  You are
8842responsible to make sure a string is pre-extended long enough to
8843receive any result that might be written into a string.  You can't use a
8844string literal (or other read-only string) as an argument to
8845L<C<syscall>|/syscall NUMBER, LIST> because Perl has to assume that any
8846string pointer might be written through.  If your
8847integer arguments are not literals and have never been interpreted in a
8848numeric context, you may need to add C<0> to them to force them to look
8849like numbers.  This emulates the
8850L<C<syswrite>|/syswrite FILEHANDLE,SCALAR,LENGTH,OFFSET> function (or
8851vice versa):
8852
8853    require 'syscall.ph';        # may need to run h2ph
8854    my $s = "hi there\n";
8855    syscall(SYS_write(), fileno(STDOUT), $s, length $s);
8856
8857Note that Perl supports passing of up to only 14 arguments to your syscall,
8858which in practice should (usually) suffice.
8859
8860Syscall returns whatever value returned by the system call it calls.
8861If the system call fails, L<C<syscall>|/syscall NUMBER, LIST> returns
8862C<-1> and sets L<C<$!>|perlvar/$!> (errno).
8863Note that some system calls I<can> legitimately return C<-1>.  The proper
8864way to handle such calls is to assign C<$! = 0> before the call, then
8865check the value of L<C<$!>|perlvar/$!> if
8866L<C<syscall>|/syscall NUMBER, LIST> returns C<-1>.
8867
8868There's a problem with C<syscall(SYS_pipe())>: it returns the file
8869number of the read end of the pipe it creates, but there is no way
8870to retrieve the file number of the other end.  You can avoid this
8871problem by using L<C<pipe>|/pipe READHANDLE,WRITEHANDLE> instead.
8872
8873Portability issues: L<perlport/syscall>.
8874
8875=item sysopen FILEHANDLE,FILENAME,MODE
8876X<sysopen>
8877
8878=item sysopen FILEHANDLE,FILENAME,MODE,PERMS
8879
8880=for Pod::Functions +5.002 open a file, pipe, or descriptor
8881
8882Opens the file whose filename is given by FILENAME, and associates it with
8883FILEHANDLE.  If FILEHANDLE is an expression, its value is used as the real
8884filehandle wanted; an undefined scalar will be suitably autovivified.  This
8885function calls the underlying operating system's L<open(2)> function with the
8886parameters FILENAME, MODE, and PERMS.
8887
8888Returns true on success and L<C<undef>|/undef EXPR> otherwise.
8889
8890L<PerlIO> layers will be applied to the handle the same way they would in an
8891L<C<open>|/open FILEHANDLE,MODE,EXPR> call that does not specify layers. That is,
8892the current value of L<C<${^OPEN}>|perlvar/${^OPEN}> as set by the L<open>
8893pragma in a lexical scope, or the C<-C> commandline option or C<PERL_UNICODE>
8894environment variable in the main program scope, falling back to the platform
8895defaults as described in L<PerlIO/Defaults and how to override them>. If you
8896want to remove any layers that may transform the byte stream, use
8897L<C<binmode>|/binmode FILEHANDLE, LAYER> after opening it.
8898
8899The possible values and flag bits of the MODE parameter are
8900system-dependent; they are available via the standard module
8901L<C<Fcntl>|Fcntl>.  See the documentation of your operating system's
8902L<open(2)> syscall to see
8903which values and flag bits are available.  You may combine several flags
8904using the C<|>-operator.
8905
8906Some of the most common values are C<O_RDONLY> for opening the file in
8907read-only mode, C<O_WRONLY> for opening the file in write-only mode,
8908and C<O_RDWR> for opening the file in read-write mode.
8909X<O_RDONLY> X<O_RDWR> X<O_WRONLY>
8910
8911For historical reasons, some values work on almost every system
8912supported by Perl: 0 means read-only, 1 means write-only, and 2
8913means read/write.  We know that these values do I<not> work under
8914OS/390; you probably don't want to use them in new code.
8915
8916If the file named by FILENAME does not exist and the
8917L<C<open>|/open FILEHANDLE,MODE,EXPR> call creates
8918it (typically because MODE includes the C<O_CREAT> flag), then the value of
8919PERMS specifies the permissions of the newly created file.  If you omit
8920the PERMS argument to L<C<sysopen>|/sysopen FILEHANDLE,FILENAME,MODE>,
8921Perl uses the octal value C<0666>.
8922These permission values need to be in octal, and are modified by your
8923process's current L<C<umask>|/umask EXPR>.
8924X<O_CREAT>
8925
8926In many systems the C<O_EXCL> flag is available for opening files in
8927exclusive mode.  This is B<not> locking: exclusiveness means here that
8928if the file already exists,
8929L<C<sysopen>|/sysopen FILEHANDLE,FILENAME,MODE> fails.  C<O_EXCL> may
8930not work
8931on network filesystems, and has no effect unless the C<O_CREAT> flag
8932is set as well.  Setting C<O_CREAT|O_EXCL> prevents the file from
8933being opened if it is a symbolic link.  It does not protect against
8934symbolic links in the file's path.
8935X<O_EXCL>
8936
8937Sometimes you may want to truncate an already-existing file.  This
8938can be done using the C<O_TRUNC> flag.  The behavior of
8939C<O_TRUNC> with C<O_RDONLY> is undefined.
8940X<O_TRUNC>
8941
8942You should seldom if ever use C<0644> as argument to
8943L<C<sysopen>|/sysopen FILEHANDLE,FILENAME,MODE>, because
8944that takes away the user's option to have a more permissive umask.
8945Better to omit it.  See L<C<umask>|/umask EXPR> for more on this.
8946
8947This function has no direct relation to the usage of
8948L<C<sysread>|/sysread FILEHANDLE,SCALAR,LENGTH,OFFSET>,
8949L<C<syswrite>|/syswrite FILEHANDLE,SCALAR,LENGTH,OFFSET>,
8950or L<C<sysseek>|/sysseek FILEHANDLE,POSITION,WHENCE>.  A handle opened with
8951this function can be used with buffered IO just as one opened with
8952L<C<open>|/open FILEHANDLE,MODE,EXPR> can be used with unbuffered IO.
8953
8954Note that under Perls older than 5.8.0,
8955L<C<sysopen>|/sysopen FILEHANDLE,FILENAME,MODE> depends on the
8956L<fdopen(3)> C library function.  On many Unix systems, L<fdopen(3)> is known
8957to fail when file descriptors exceed a certain value, typically 255.  If
8958you need more file descriptors than that, consider using the
8959L<C<POSIX::open>|POSIX/C<open>> function.  For Perls 5.8.0 and later,
8960PerlIO is (most often) the default.
8961
8962See L<perlopentut> for a kinder, gentler explanation of opening files.
8963
8964Portability issues: L<perlport/sysopen>.
8965
8966=item sysread FILEHANDLE,SCALAR,LENGTH,OFFSET
8967X<sysread>
8968
8969=item sysread FILEHANDLE,SCALAR,LENGTH
8970
8971=for Pod::Functions fixed-length unbuffered input from a filehandle
8972
8973Attempts to read LENGTH bytes of data into variable SCALAR from the
8974specified FILEHANDLE, using L<read(2)>.  It bypasses any L<PerlIO> layers
8975including buffered IO (but is affected by the presence of the C<:utf8>
8976layer as described later), so mixing this with other kinds of reads,
8977L<C<print>|/print FILEHANDLE LIST>, L<C<write>|/write FILEHANDLE>,
8978L<C<seek>|/seek FILEHANDLE,POSITION,WHENCE>,
8979L<C<tell>|/tell FILEHANDLE>, or L<C<eof>|/eof FILEHANDLE> can cause
8980confusion because the
8981C<:perlio> or C<:crlf> layers usually buffer data.  Returns the number of
8982bytes actually read, C<0> at end of file, or undef if there was an
8983error (in the latter case L<C<$!>|perlvar/$!> is also set).  SCALAR will
8984be grown or
8985shrunk so that the last byte actually read is the last byte of the
8986scalar after the read.
8987
8988An OFFSET may be specified to place the read data at some place in the
8989string other than the beginning.  A negative OFFSET specifies
8990placement at that many characters counting backwards from the end of
8991the string.  A positive OFFSET greater than the length of SCALAR
8992results in the string being padded to the required size with C<"\0">
8993bytes before the result of the read is appended.
8994
8995There is no syseof() function, which is ok, since
8996L<C<eof>|/eof FILEHANDLE> doesn't work well on device files (like ttys)
8997anyway.  Use L<C<sysread>|/sysread FILEHANDLE,SCALAR,LENGTH,OFFSET> and
8998check for a return value of 0 to decide whether you're done.
8999
9000Note that if the filehandle has been marked as C<:utf8>, C<sysread> will
9001throw an exception.  The C<:encoding(...)> layer implicitly
9002introduces the C<:utf8> layer.  See
9003L<C<binmode>|/binmode FILEHANDLE, LAYER>,
9004L<C<open>|/open FILEHANDLE,MODE,EXPR>, and the L<open> pragma.
9005
9006=item sysseek FILEHANDLE,POSITION,WHENCE
9007X<sysseek> X<lseek>
9008
9009=for Pod::Functions +5.004 position I/O pointer on handle used with sysread and syswrite
9010
9011Sets FILEHANDLE's system position I<in bytes> using L<lseek(2)>.  FILEHANDLE may
9012be an expression whose value gives the name of the filehandle.  The values
9013for WHENCE are C<0> to set the new position to POSITION; C<1> to set it
9014to the current position plus POSITION; and C<2> to set it to EOF plus
9015POSITION, typically negative.
9016
9017Note the emphasis on bytes: even if the filehandle has been set to operate
9018on characters (for example using the C<:encoding(UTF-8)> I/O layer), the
9019L<C<seek>|/seek FILEHANDLE,POSITION,WHENCE>,
9020L<C<tell>|/tell FILEHANDLE>, and
9021L<C<sysseek>|/sysseek FILEHANDLE,POSITION,WHENCE>
9022family of functions use byte offsets, not character offsets,
9023because seeking to a character offset would be very slow in a UTF-8 file.
9024
9025L<C<sysseek>|/sysseek FILEHANDLE,POSITION,WHENCE> bypasses normal
9026buffered IO, so mixing it with reads other than
9027L<C<sysread>|/sysread FILEHANDLE,SCALAR,LENGTH,OFFSET> (for example
9028L<C<readline>|/readline EXPR> or
9029L<C<read>|/read FILEHANDLE,SCALAR,LENGTH,OFFSET>),
9030L<C<print>|/print FILEHANDLE LIST>, L<C<write>|/write FILEHANDLE>,
9031L<C<seek>|/seek FILEHANDLE,POSITION,WHENCE>,
9032L<C<tell>|/tell FILEHANDLE>, or L<C<eof>|/eof FILEHANDLE> may cause
9033confusion.
9034
9035For WHENCE, you may also use the constants C<SEEK_SET>, C<SEEK_CUR>,
9036and C<SEEK_END> (start of the file, current position, end of the file)
9037from the L<Fcntl> module.  Use of the constants is also more portable
9038than relying on 0, 1, and 2.  For example to define a "systell" function:
9039
9040    use Fcntl 'SEEK_CUR';
9041    sub systell { sysseek($_[0], 0, SEEK_CUR) }
9042
9043Returns the new position, or the undefined value on failure.  A position
9044of zero is returned as the string C<"0 but true">; thus
9045L<C<sysseek>|/sysseek FILEHANDLE,POSITION,WHENCE> returns
9046true on success and false on failure, yet you can still easily determine
9047the new position.
9048
9049=item system LIST
9050X<system> X<shell>
9051
9052=item system PROGRAM LIST
9053
9054=for Pod::Functions run a separate program
9055
9056Does exactly the same thing as L<C<exec>|/exec LIST>, except that a fork is
9057done first and the parent process waits for the child process to
9058exit.  Note that argument processing varies depending on the
9059number of arguments.  If there is more than one argument in LIST,
9060or if LIST is an array with more than one value, starts the program
9061given by the first element of the list with arguments given by the
9062rest of the list.  If there is only one scalar argument, the argument
9063is checked for shell metacharacters, and if there are any, the
9064entire argument is passed to the system's command shell for parsing
9065(this is C</bin/sh -c> on Unix platforms, but varies on other
9066platforms).  If there are no shell metacharacters in the argument,
9067it is split into words and passed directly to C<execvp>, which is
9068more efficient.  On Windows, only the C<system PROGRAM LIST> syntax will
9069reliably avoid using the shell; C<system LIST>, even with more than one
9070element, will fall back to the shell if the first spawn fails.
9071
9072Perl will attempt to flush all files opened for
9073output before any operation that may do a fork, but this may not be
9074supported on some platforms (see L<perlport>).  To be safe, you may need
9075to set L<C<$E<verbar>>|perlvar/$E<verbar>> (C<$AUTOFLUSH> in L<English>)
9076or call the C<autoflush> method of L<C<IO::Handle>|IO::Handle/METHODS>
9077on any open handles.
9078
9079The return value is the exit status of the program as returned by the
9080L<C<wait>|/wait> call.  To get the actual exit value, shift right by
9081eight (see below).  See also L<C<exec>|/exec LIST>.  This is I<not> what
9082you want to use to capture the output from a command; for that you
9083should use merely backticks or
9084L<C<qxE<sol>E<sol>>|/qxE<sol>STRINGE<sol>>, as described in
9085L<perlop/"`STRING`">.  Return value of -1 indicates a failure to start
9086the program or an error of the L<wait(2)> system call (inspect
9087L<C<$!>|perlvar/$!> for the reason).
9088
9089If you'd like to make L<C<system>|/system LIST> (and many other bits of
9090Perl) die on error, have a look at the L<autodie> pragma.
9091
9092Like L<C<exec>|/exec LIST>, L<C<system>|/system LIST> allows you to lie
9093to a program about its name if you use the C<system PROGRAM LIST>
9094syntax.  Again, see L<C<exec>|/exec LIST>.
9095
9096Since C<SIGINT> and C<SIGQUIT> are ignored during the execution of
9097L<C<system>|/system LIST>, if you expect your program to terminate on
9098receipt of these signals you will need to arrange to do so yourself
9099based on the return value.
9100
9101    my @args = ("command", "arg1", "arg2");
9102    system(@args) == 0
9103        or die "system @args failed: $?";
9104
9105If you'd like to manually inspect L<C<system>|/system LIST>'s failure,
9106you can check all possible failure modes by inspecting
9107L<C<$?>|perlvar/$?> like this:
9108
9109    if ($? == -1) {
9110        print "failed to execute: $!\n";
9111    }
9112    elsif ($? & 127) {
9113        printf "child died with signal %d, %s coredump\n",
9114            ($? & 127),  ($? & 128) ? 'with' : 'without';
9115    }
9116    else {
9117        printf "child exited with value %d\n", $? >> 8;
9118    }
9119
9120Alternatively, you may inspect the value of
9121L<C<${^CHILD_ERROR_NATIVE}>|perlvar/${^CHILD_ERROR_NATIVE}> with the
9122L<C<W*()>|POSIX/C<WIFEXITED>> calls from the L<POSIX> module.
9123
9124When L<C<system>|/system LIST>'s arguments are executed indirectly by
9125the shell, results and return codes are subject to its quirks.
9126See L<perlop/"`STRING`"> and L<C<exec>|/exec LIST> for details.
9127
9128Since L<C<system>|/system LIST> does a L<C<fork>|/fork> and
9129L<C<wait>|/wait> it may affect a C<SIGCHLD> handler.  See L<perlipc> for
9130details.
9131
9132Portability issues: L<perlport/system>.
9133
9134=item syswrite FILEHANDLE,SCALAR,LENGTH,OFFSET
9135X<syswrite>
9136
9137=item syswrite FILEHANDLE,SCALAR,LENGTH
9138
9139=item syswrite FILEHANDLE,SCALAR
9140
9141=for Pod::Functions fixed-length unbuffered output to a filehandle
9142
9143Attempts to write LENGTH bytes of data from variable SCALAR to the
9144specified FILEHANDLE, using L<write(2)>.  If LENGTH is
9145not specified, writes whole SCALAR.  It bypasses any L<PerlIO> layers
9146including buffered IO (but is affected by the presence of the C<:utf8>
9147layer as described later), so
9148mixing this with reads (other than C<sysread)>),
9149L<C<print>|/print FILEHANDLE LIST>, L<C<write>|/write FILEHANDLE>,
9150L<C<seek>|/seek FILEHANDLE,POSITION,WHENCE>,
9151L<C<tell>|/tell FILEHANDLE>, or L<C<eof>|/eof FILEHANDLE> may cause
9152confusion because the C<:perlio> and C<:crlf> layers usually buffer data.
9153Returns the number of bytes actually written, or L<C<undef>|/undef EXPR>
9154if there was an error (in this case the errno variable
9155L<C<$!>|perlvar/$!> is also set).  If the LENGTH is greater than the
9156data available in the SCALAR after the OFFSET, only as much data as is
9157available will be written.
9158
9159An OFFSET may be specified to write the data from some part of the
9160string other than the beginning.  A negative OFFSET specifies writing
9161that many characters counting backwards from the end of the string.
9162If SCALAR is of length zero, you can only use an OFFSET of 0.
9163
9164B<WARNING>: If the filehandle is marked C<:utf8>, C<syswrite> will raise an exception.
9165The C<:encoding(...)> layer implicitly introduces the C<:utf8> layer.
9166Alternately, if the handle is not marked with an encoding but you
9167attempt to write characters with code points over 255, raises an exception.
9168See L<C<binmode>|/binmode FILEHANDLE, LAYER>,
9169L<C<open>|/open FILEHANDLE,MODE,EXPR>, and the L<open> pragma.
9170
9171=item tell FILEHANDLE
9172X<tell>
9173
9174=item tell
9175
9176=for Pod::Functions get current seekpointer on a filehandle
9177
9178Returns the current position I<in bytes> for FILEHANDLE, or -1 on
9179error.  FILEHANDLE may be an expression whose value gives the name of
9180the actual filehandle.  If FILEHANDLE is omitted, assumes the file
9181last read.
9182
9183Note the emphasis on bytes: even if the filehandle has been set to operate
9184on characters (for example using the C<:encoding(UTF-8)> I/O layer), the
9185L<C<seek>|/seek FILEHANDLE,POSITION,WHENCE>,
9186L<C<tell>|/tell FILEHANDLE>, and
9187L<C<sysseek>|/sysseek FILEHANDLE,POSITION,WHENCE>
9188family of functions use byte offsets, not character offsets,
9189because seeking to a character offset would be very slow in a UTF-8 file.
9190
9191The return value of L<C<tell>|/tell FILEHANDLE> for the standard streams
9192like the STDIN depends on the operating system: it may return -1 or
9193something else.  L<C<tell>|/tell FILEHANDLE> on pipes, fifos, and
9194sockets usually returns -1.
9195
9196There is no C<systell> function.  Use
9197L<C<sysseek($fh, 0, 1)>|/sysseek FILEHANDLE,POSITION,WHENCE> for that.
9198
9199Do not use L<C<tell>|/tell FILEHANDLE> (or other buffered I/O
9200operations) on a filehandle that has been manipulated by
9201L<C<sysread>|/sysread FILEHANDLE,SCALAR,LENGTH,OFFSET>,
9202L<C<syswrite>|/syswrite FILEHANDLE,SCALAR,LENGTH,OFFSET>, or
9203L<C<sysseek>|/sysseek FILEHANDLE,POSITION,WHENCE>.  Those functions
9204ignore the buffering, while L<C<tell>|/tell FILEHANDLE> does not.
9205
9206=item telldir DIRHANDLE
9207X<telldir>
9208
9209=for Pod::Functions get current seekpointer on a directory handle
9210
9211Returns the current position of the L<C<readdir>|/readdir DIRHANDLE>
9212routines on DIRHANDLE.  Value may be given to
9213L<C<seekdir>|/seekdir DIRHANDLE,POS> to access a particular location in
9214a directory.  L<C<telldir>|/telldir DIRHANDLE> has the same caveats
9215about possible directory compaction as the corresponding system library
9216routine.
9217
9218=item tie VARIABLE,CLASSNAME,LIST
9219X<tie>
9220
9221=for Pod::Functions +5.002 bind a variable to an object class
9222
9223This function binds a variable to a package class that will provide the
9224implementation for the variable.  VARIABLE is the name of the variable
9225to be enchanted.  CLASSNAME is the name of a class implementing objects
9226of correct type.  Any additional arguments are passed to the
9227appropriate constructor
9228method of the class (meaning C<TIESCALAR>, C<TIEHANDLE>, C<TIEARRAY>,
9229or C<TIEHASH>).  Typically these are arguments such as might be passed
9230to the L<dbm_open(3)> function of C.  The object returned by the
9231constructor is also returned by the
9232L<C<tie>|/tie VARIABLE,CLASSNAME,LIST> function, which would be useful
9233if you want to access other methods in CLASSNAME.
9234
9235Note that functions such as L<C<keys>|/keys HASH> and
9236L<C<values>|/values HASH> may return huge lists when used on large
9237objects, like DBM files.  You may prefer to use the L<C<each>|/each
9238HASH> function to iterate over such.  Example:
9239
9240    # print out history file offsets
9241    use NDBM_File;
9242    tie(my %HIST, 'NDBM_File', '/usr/lib/news/history', 1, 0);
9243    while (my ($key,$val) = each %HIST) {
9244        print $key, ' = ', unpack('L', $val), "\n";
9245    }
9246
9247A class implementing a hash should have the following methods:
9248
9249    TIEHASH classname, LIST
9250    FETCH this, key
9251    STORE this, key, value
9252    DELETE this, key
9253    CLEAR this
9254    EXISTS this, key
9255    FIRSTKEY this
9256    NEXTKEY this, lastkey
9257    SCALAR this
9258    DESTROY this
9259    UNTIE this
9260
9261A class implementing an ordinary array should have the following methods:
9262
9263    TIEARRAY classname, LIST
9264    FETCH this, key
9265    STORE this, key, value
9266    FETCHSIZE this
9267    STORESIZE this, count
9268    CLEAR this
9269    PUSH this, LIST
9270    POP this
9271    SHIFT this
9272    UNSHIFT this, LIST
9273    SPLICE this, offset, length, LIST
9274    EXTEND this, count
9275    DELETE this, key
9276    EXISTS this, key
9277    DESTROY this
9278    UNTIE this
9279
9280A class implementing a filehandle should have the following methods:
9281
9282    TIEHANDLE classname, LIST
9283    READ this, scalar, length, offset
9284    READLINE this
9285    GETC this
9286    WRITE this, scalar, length, offset
9287    PRINT this, LIST
9288    PRINTF this, format, LIST
9289    BINMODE this
9290    EOF this
9291    FILENO this
9292    SEEK this, position, whence
9293    TELL this
9294    OPEN this, mode, LIST
9295    CLOSE this
9296    DESTROY this
9297    UNTIE this
9298
9299A class implementing a scalar should have the following methods:
9300
9301    TIESCALAR classname, LIST
9302    FETCH this,
9303    STORE this, value
9304    DESTROY this
9305    UNTIE this
9306
9307Not all methods indicated above need be implemented.  See L<perltie>,
9308L<Tie::Hash>, L<Tie::Array>, L<Tie::Scalar>, and L<Tie::Handle>.
9309
9310Unlike L<C<dbmopen>|/dbmopen HASH,DBNAME,MASK>, the
9311L<C<tie>|/tie VARIABLE,CLASSNAME,LIST> function will not
9312L<C<use>|/use Module VERSION LIST> or L<C<require>|/require VERSION> a
9313module for you; you need to do that explicitly yourself.  See L<DB_File>
9314or the L<Config> module for interesting
9315L<C<tie>|/tie VARIABLE,CLASSNAME,LIST> implementations.
9316
9317For further details see L<perltie>, L<C<tied>|/tied VARIABLE>.
9318
9319=item tied VARIABLE
9320X<tied>
9321
9322=for Pod::Functions get a reference to the object underlying a tied variable
9323
9324Returns a reference to the object underlying VARIABLE (the same value
9325that was originally returned by the
9326L<C<tie>|/tie VARIABLE,CLASSNAME,LIST> call that bound the variable
9327to a package.)  Returns the undefined value if VARIABLE isn't tied to a
9328package.
9329
9330=item time
9331X<time> X<epoch>
9332
9333=for Pod::Functions return number of seconds since 1970
9334
9335Returns the number of non-leap seconds since whatever time the system
9336considers to be the epoch, suitable for feeding to
9337L<C<gmtime>|/gmtime EXPR> and L<C<localtime>|/localtime EXPR>.  On most
9338systems the epoch is 00:00:00 UTC, January 1, 1970;
9339a prominent exception being Mac OS Classic which uses 00:00:00, January 1,
93401904 in the current local time zone for its epoch.
9341
9342For measuring time in better granularity than one second, use the
9343L<Time::HiRes> module from Perl 5.8 onwards (or from CPAN before then), or,
9344if you have L<gettimeofday(2)>, you may be able to use the
9345L<C<syscall>|/syscall NUMBER, LIST> interface of Perl.  See L<perlfaq8>
9346for details.
9347
9348For date and time processing look at the many related modules on CPAN.
9349For a comprehensive date and time representation look at the
9350L<DateTime> module.
9351
9352=item times
9353X<times>
9354
9355=for Pod::Functions return elapsed time for self and child processes
9356
9357Returns a four-element list giving the user and system times in
9358seconds for this process and any exited children of this process.
9359
9360    my ($user,$system,$cuser,$csystem) = times;
9361
9362In scalar context, L<C<times>|/times> returns C<$user>.
9363
9364Children's times are only included for terminated children.
9365
9366Portability issues: L<perlport/times>.
9367
9368=item tr///
9369
9370=for Pod::Functions transliterate a string
9371
9372The transliteration operator.  Same as
9373L<C<yE<sol>E<sol>E<sol>>|/yE<sol>E<sol>E<sol>>.  See
9374L<perlop/"Quote-Like Operators">.
9375
9376=item truncate FILEHANDLE,LENGTH
9377X<truncate>
9378
9379=item truncate EXPR,LENGTH
9380
9381=for Pod::Functions shorten a file
9382
9383Truncates the file opened on FILEHANDLE, or named by EXPR, to the
9384specified length.  Raises an exception if truncate isn't implemented
9385on your system.  Returns true if successful, L<C<undef>|/undef EXPR> on
9386error.
9387
9388The behavior is undefined if LENGTH is greater than the length of the
9389file.
9390
9391The position in the file of FILEHANDLE is left unchanged.  You may want to
9392call L<seek|/"seek FILEHANDLE,POSITION,WHENCE"> before writing to the
9393file.
9394
9395Portability issues: L<perlport/truncate>.
9396
9397=item uc EXPR
9398X<uc> X<uppercase> X<toupper>
9399
9400=item uc
9401
9402=for Pod::Functions return upper-case version of a string
9403
9404Returns an uppercased version of EXPR.  If EXPR is omitted, uses
9405L<C<$_>|perlvar/$_>.
9406
9407    my $str = uc("Perl is GREAT"); # "PERL IS GREAT"
9408
9409This function behaves the same way under various pragmas, such as in a locale,
9410as L<C<lc>|/lc EXPR> does.
9411
9412If you want titlecase mapping on initial letters see
9413L<C<ucfirst>|/ucfirst EXPR> instead.
9414
9415B<Note:> This is the internal function implementing the
9416L<C<\U>|perlop/"Quote and Quote-like Operators"> escape in double-quoted
9417strings.
9418
9419    my $str = "Perl is \Ugreat\E"; # "Perl is GREAT"
9420
9421=item ucfirst EXPR
9422X<ucfirst> X<uppercase>
9423
9424=item ucfirst
9425
9426=for Pod::Functions return a string with just the next letter in upper case
9427
9428Returns the value of EXPR with the first character in uppercase
9429(titlecase in Unicode).  This is the internal function implementing
9430the C<\u> escape in double-quoted strings.
9431
9432If EXPR is omitted, uses L<C<$_>|perlvar/$_>.
9433
9434This function behaves the same way under various pragmas, such as in a locale,
9435as L<C<lc>|/lc EXPR> does.
9436
9437=item umask EXPR
9438X<umask>
9439
9440=item umask
9441
9442=for Pod::Functions set file creation mode mask
9443
9444Sets the umask for the process to EXPR and returns the previous value.
9445If EXPR is omitted, merely returns the current umask.
9446
9447The Unix permission C<rwxr-x---> is represented as three sets of three
9448bits, or three octal digits: C<0750> (the leading 0 indicates octal
9449and isn't one of the digits).  The L<C<umask>|/umask EXPR> value is such
9450a number representing disabled permissions bits.  The permission (or
9451"mode") values you pass L<C<mkdir>|/mkdir FILENAME,MODE> or
9452L<C<sysopen>|/sysopen FILEHANDLE,FILENAME,MODE> are modified by your
9453umask, so even if you tell
9454L<C<sysopen>|/sysopen FILEHANDLE,FILENAME,MODE> to create a file with
9455permissions C<0777>, if your umask is C<0022>, then the file will
9456actually be created with permissions C<0755>.  If your
9457L<C<umask>|/umask EXPR> were C<0027> (group can't write; others can't
9458read, write, or execute), then passing
9459L<C<sysopen>|/sysopen FILEHANDLE,FILENAME,MODE> C<0666> would create a
9460file with mode C<0640> (because C<0666 &~ 027> is C<0640>).
9461
9462Here's some advice: supply a creation mode of C<0666> for regular
9463files (in L<C<sysopen>|/sysopen FILEHANDLE,FILENAME,MODE>) and one of
9464C<0777> for directories (in L<C<mkdir>|/mkdir FILENAME,MODE>) and
9465executable files.  This gives users the freedom of
9466choice: if they want protected files, they might choose process umasks
9467of C<022>, C<027>, or even the particularly antisocial mask of C<077>.
9468Programs should rarely if ever make policy decisions better left to
9469the user.  The exception to this is when writing files that should be
9470kept private: mail files, web browser cookies, F<.rhosts> files, and
9471so on.
9472
9473If L<umask(2)> is not implemented on your system and you are trying to
9474restrict access for I<yourself> (i.e., C<< (EXPR & 0700) > 0 >>),
9475raises an exception.  If L<umask(2)> is not implemented and you are
9476not trying to restrict access for yourself, returns
9477L<C<undef>|/undef EXPR>.
9478
9479Remember that a umask is a number, usually given in octal; it is I<not> a
9480string of octal digits.  See also L<C<oct>|/oct EXPR>, if all you have
9481is a string.
9482
9483Portability issues: L<perlport/umask>.
9484
9485=item undef EXPR
9486X<undef> X<undefine>
9487
9488=item undef
9489
9490=for Pod::Functions remove a variable or function definition
9491
9492Undefines the value of EXPR, which must be an lvalue.  Use only on a
9493scalar value, an array (using C<@>), a hash (using C<%>), a subroutine
9494(using C<&>), or a typeglob (using C<*>).  Saying C<undef $hash{$key}>
9495will probably not do what you expect on most predefined variables or
9496DBM list values, so don't do that; see L<C<delete>|/delete EXPR>.
9497Always returns the undefined value.
9498You can omit the EXPR, in which case nothing is
9499undefined, but you still get an undefined value that you could, for
9500instance, return from a subroutine, assign to a variable, or pass as a
9501parameter.  Examples:
9502
9503    undef $foo;
9504    undef $bar{'blurfl'};      # Compare to: delete $bar{'blurfl'};
9505    undef @ary;
9506    undef %hash;
9507    undef &mysub;
9508    undef *xyz;       # destroys $xyz, @xyz, %xyz, &xyz, etc.
9509    return (wantarray ? (undef, $errmsg) : undef) if $they_blew_it;
9510    select undef, undef, undef, 0.25;
9511    my ($x, $y, undef, $z) = foo();    # Ignore third value returned
9512
9513Note that this is a unary operator, not a list operator.
9514
9515=item unlink LIST
9516X<unlink> X<delete> X<remove> X<rm> X<del>
9517
9518=item unlink
9519
9520=for Pod::Functions remove one link to a file
9521
9522Deletes a list of files.  On success, it returns the number of files
9523it successfully deleted.  On failure, it returns false and sets
9524L<C<$!>|perlvar/$!> (errno):
9525
9526    my $unlinked = unlink 'a', 'b', 'c';
9527    unlink @goners;
9528    unlink glob "*.bak";
9529
9530On error, L<C<unlink>|/unlink LIST> will not tell you which files it
9531could not remove.
9532If you want to know which files you could not remove, try them one
9533at a time:
9534
9535     foreach my $file ( @goners ) {
9536         unlink $file or warn "Could not unlink $file: $!";
9537     }
9538
9539Note: L<C<unlink>|/unlink LIST> will not attempt to delete directories
9540unless you are
9541superuser and the B<-U> flag is supplied to Perl.  Even if these
9542conditions are met, be warned that unlinking a directory can inflict
9543damage on your filesystem.  Finally, using L<C<unlink>|/unlink LIST> on
9544directories is not supported on many operating systems.  Use
9545L<C<rmdir>|/rmdir FILENAME> instead.
9546
9547If LIST is omitted, L<C<unlink>|/unlink LIST> uses L<C<$_>|perlvar/$_>.
9548
9549=item unpack TEMPLATE,EXPR
9550X<unpack>
9551
9552=item unpack TEMPLATE
9553
9554=for Pod::Functions convert binary structure into normal perl variables
9555
9556L<C<unpack>|/unpack TEMPLATE,EXPR> does the reverse of
9557L<C<pack>|/pack TEMPLATE,LIST>: it takes a string
9558and expands it out into a list of values.
9559(In scalar context, it returns merely the first value produced.)
9560
9561If EXPR is omitted, unpacks the L<C<$_>|perlvar/$_> string.
9562See L<perlpacktut> for an introduction to this function.
9563
9564The string is broken into chunks described by the TEMPLATE.  Each chunk
9565is converted separately to a value.  Typically, either the string is a result
9566of L<C<pack>|/pack TEMPLATE,LIST>, or the characters of the string
9567represent a C structure of some kind.
9568
9569The TEMPLATE has the same format as in the
9570L<C<pack>|/pack TEMPLATE,LIST> function.
9571Here's a subroutine that does substring:
9572
9573    sub substr {
9574        my ($what, $where, $howmuch) = @_;
9575        unpack("x$where a$howmuch", $what);
9576    }
9577
9578and then there's
9579
9580    sub ordinal { unpack("W",$_[0]); } # same as ord()
9581
9582In addition to fields allowed in L<C<pack>|/pack TEMPLATE,LIST>, you may
9583prefix a field with a %<number> to indicate that
9584you want a <number>-bit checksum of the items instead of the items
9585themselves.  Default is a 16-bit checksum.  The checksum is calculated by
9586summing numeric values of expanded values (for string fields the sum of
9587C<ord($char)> is taken; for bit fields the sum of zeroes and ones).
9588
9589For example, the following
9590computes the same number as the System V sum program:
9591
9592    my $checksum = do {
9593        local $/;  # slurp!
9594        unpack("%32W*", readline) % 65535;
9595    };
9596
9597The following efficiently counts the number of set bits in a bit vector:
9598
9599    my $setbits = unpack("%32b*", $selectmask);
9600
9601The C<p> and C<P> formats should be used with care.  Since Perl
9602has no way of checking whether the value passed to
9603L<C<unpack>|/unpack TEMPLATE,EXPR>
9604corresponds to a valid memory location, passing a pointer value that's
9605not known to be valid is likely to have disastrous consequences.
9606
9607If there are more pack codes or if the repeat count of a field or a group
9608is larger than what the remainder of the input string allows, the result
9609is not well defined: the repeat count may be decreased, or
9610L<C<unpack>|/unpack TEMPLATE,EXPR> may produce empty strings or zeros,
9611or it may raise an exception.
9612If the input string is longer than one described by the TEMPLATE,
9613the remainder of that input string is ignored.
9614
9615See L<C<pack>|/pack TEMPLATE,LIST> for more examples and notes.
9616
9617=item unshift ARRAY,LIST
9618X<unshift>
9619
9620=for Pod::Functions prepend more elements to the beginning of a list
9621
9622Does the opposite of a L<C<shift>|/shift ARRAY>.  Or the opposite of a
9623L<C<push>|/push ARRAY,LIST>,
9624depending on how you look at it.  Prepends list to the front of the
9625array and returns the new number of elements in the array.
9626
9627    unshift(@ARGV, '-e') unless $ARGV[0] =~ /^-/;
9628
9629Note the LIST is prepended whole, not one element at a time, so the
9630prepended elements stay in the same order.  Use
9631L<C<reverse>|/reverse LIST> to do the reverse.
9632
9633Starting with Perl 5.14, an experimental feature allowed
9634L<C<unshift>|/unshift ARRAY,LIST> to take
9635a scalar expression. This experiment has been deemed unsuccessful, and was
9636removed as of Perl 5.24.
9637
9638=item untie VARIABLE
9639X<untie>
9640
9641=for Pod::Functions break a tie binding to a variable
9642
9643Breaks the binding between a variable and a package.
9644(See L<tie|/tie VARIABLE,CLASSNAME,LIST>.)
9645Has no effect if the variable is not tied.
9646
9647=item use Module VERSION LIST
9648X<use> X<module> X<import>
9649
9650=item use Module VERSION
9651
9652=item use Module LIST
9653
9654=item use Module
9655
9656=for Pod::Functions load in a module at compile time and import its namespace
9657
9658Imports some semantics into the current package from the named module,
9659generally by aliasing certain subroutine or variable names into your
9660package.  It is exactly equivalent to
9661
9662    BEGIN { require Module; Module->import( LIST ); }
9663
9664except that Module I<must> be a bareword.
9665The importation can be made conditional by using the L<if> module.
9666
9667The C<BEGIN> forces the L<C<require>|/require VERSION> and
9668L<C<import>|/import LIST> to happen at compile time.  The
9669L<C<require>|/require VERSION> makes sure the module is loaded into
9670memory if it hasn't been yet.  The L<C<import>|/import LIST> is not a
9671builtin; it's just an ordinary static method
9672call into the C<Module> package to tell the module to import the list of
9673features back into the current package.  The module can implement its
9674L<C<import>|/import LIST> method any way it likes, though most modules
9675just choose to derive their L<C<import>|/import LIST> method via
9676inheritance from the C<Exporter> class that is defined in the
9677L<C<Exporter>|Exporter> module.  See L<Exporter>.  If no
9678L<C<import>|/import LIST> method can be found, then the call is skipped,
9679even if there is an AUTOLOAD method.
9680
9681If you do not want to call the package's L<C<import>|/import LIST>
9682method (for instance,
9683to stop your namespace from being altered), explicitly supply the empty list:
9684
9685    use Module ();
9686
9687That is exactly equivalent to
9688
9689    BEGIN { require Module }
9690
9691If the VERSION argument is present between Module and LIST, then the
9692L<C<use>|/use Module VERSION LIST> will call the C<VERSION> method in
9693class Module with the given version as an argument:
9694
9695    use Module 12.34;
9696
9697is equivalent to:
9698
9699    BEGIN { require Module; Module->VERSION(12.34) }
9700
9701The L<default C<VERSION> method|UNIVERSAL/C<VERSION ( [ REQUIRE ] )>>,
9702inherited from the L<C<UNIVERSAL>|UNIVERSAL> class, croaks if the given
9703version is larger than the value of the variable C<$Module::VERSION>.
9704
9705The VERSION argument cannot be an arbitrary expression.  It only counts
9706as a VERSION argument if it is a version number literal, starting with
9707either a digit or C<v> followed by a digit.  Anything that doesn't
9708look like a version literal will be parsed as the start of the LIST.
9709Nevertheless, many attempts to use an arbitrary expression as a VERSION
9710argument will appear to work, because L<Exporter>'s C<import> method
9711handles numeric arguments specially, performing version checks rather
9712than treating them as things to export.
9713
9714Again, there is a distinction between omitting LIST (L<C<import>|/import
9715LIST> called with no arguments) and an explicit empty LIST C<()>
9716(L<C<import>|/import LIST> not called).  Note that there is no comma
9717after VERSION!
9718
9719Because this is a wide-open interface, pragmas (compiler directives)
9720are also implemented this way.  Some of the currently implemented
9721pragmas are:
9722
9723    use constant;
9724    use diagnostics;
9725    use integer;
9726    use sigtrap  qw(SEGV BUS);
9727    use strict   qw(subs vars refs);
9728    use subs     qw(afunc blurfl);
9729    use warnings qw(all);
9730    use sort     qw(stable);
9731
9732Some of these pseudo-modules import semantics into the current
9733block scope (like L<C<strict>|strict> or L<C<integer>|integer>, unlike
9734ordinary modules, which import symbols into the current package (which
9735are effective through the end of the file).
9736
9737Because L<C<use>|/use Module VERSION LIST> takes effect at compile time,
9738it doesn't respect the ordinary flow control of the code being compiled.
9739In particular, putting a L<C<use>|/use Module VERSION LIST> inside the
9740false branch of a conditional doesn't prevent it
9741from being processed.  If a module or pragma only needs to be loaded
9742conditionally, this can be done using the L<if> pragma:
9743
9744    use if $] < 5.008, "utf8";
9745    use if WANT_WARNINGS, warnings => qw(all);
9746
9747There's a corresponding L<C<no>|/no MODULE VERSION LIST> declaration
9748that unimports meanings imported by L<C<use>|/use Module VERSION LIST>,
9749i.e., it calls C<< Module->unimport(LIST) >> instead of
9750L<C<import>|/import LIST>.  It behaves just as L<C<import>|/import LIST>
9751does with VERSION, an omitted or empty LIST,
9752or no unimport method being found.
9753
9754    no integer;
9755    no strict 'refs';
9756    no warnings;
9757
9758See L<perlmodlib> for a list of standard modules and pragmas.  See
9759L<perlrun|perlrun/-m[-]module> for the C<-M> and C<-m> command-line
9760options to Perl that give L<C<use>|/use Module VERSION LIST>
9761functionality from the command-line.
9762
9763=item use VERSION
9764
9765=for Pod::Functions enable Perl language features and declare required version
9766
9767Lexically enables all features available in the requested version as
9768defined by the L<feature> pragma, disabling any features not in the
9769requested version's feature bundle.  See L<feature>.
9770
9771VERSION may be either a v-string such as v5.24.1, which will be compared
9772to L<C<$^V>|perlvar/$^V> (aka $PERL_VERSION), or a numeric argument of the
9773form 5.024001, which will be compared to L<C<$]>|perlvar/$]>.  An
9774exception is raised if VERSION is greater than the version of the current
9775Perl interpreter; Perl will not attempt to parse the rest of the file.
9776Compare with L<C<require>|/require VERSION>, which can do a similar check
9777at run time.
9778
9779If the specified Perl version is 5.12 or higher, strictures are enabled
9780lexically as with L<C<use strict>|strict>.  Similarly, if the specified
9781Perl version is 5.35.0 or higher, L<warnings> are enabled.  Later use of
9782C<use VERSION> will override all behavior of a previous C<use VERSION>,
9783possibly removing the C<strict>, C<warnings>, and C<feature> added by it.
9784C<use VERSION> does not load the F<feature.pm>, F<strict.pm>, or
9785F<warnings.pm> files.
9786
9787In the current implementation, any explicit use of C<use strict> or
9788C<no strict> overrides C<use VERSION>, even if it comes before it.
9789However, this may be subject to change in a future release of Perl, so new
9790code should not rely on this fact.  It is recommended that a
9791C<use VERSION> declaration be the first significant statement within a
9792file (possibly after a C<package> statement or any amount of whitespace or
9793comment), so that its effects happen first, and other pragmata are applied
9794after it.
9795
9796Specifying VERSION as a numeric argument of the form 5.024001 should
9797generally be avoided as older less readable syntax compared to
9798v5.24.1. Before perl 5.8.0 released in 2002 the more verbose numeric
9799form was the only supported syntax, which is why you might see it in
9800older code.
9801
9802    use v5.24.1;    # compile time version check
9803    use 5.24.1;     # ditto
9804    use 5.024_001;  # ditto; older syntax compatible with perl 5.6
9805
9806This is often useful if you need to check the current Perl version before
9807L<C<use>|/use Module VERSION LIST>ing library modules that won't work
9808with older versions of Perl.
9809(We try not to do this more than we have to.)
9810
9811Symmetrically, C<no VERSION> allows you to specify that you want a version
9812of Perl older than the specified one.  Historically this was added during
9813early designs of the Raku language (formerly "Perl 6"), so that a Perl 5
9814program could begin
9815
9816    no 6;
9817
9818to declare that it is not a Perl 6 program.  As the two languages have
9819different implementations, file naming conventions, and other
9820infrastructure, this feature is now little used in practice and should be
9821avoided in newly-written code.
9822
9823Care should be taken when using the C<no VERSION> form, as it is I<only>
9824meant to be used to assert that the running Perl is of a earlier version
9825than its argument and I<not> to undo the feature-enabling side effects
9826of C<use VERSION>.
9827
9828=item utime LIST
9829X<utime>
9830
9831=for Pod::Functions set a file's last access and modify times
9832
9833Changes the access and modification times on each file of a list of
9834files.  The first two elements of the list must be the NUMERIC access
9835and modification times, in that order.  Returns the number of files
9836successfully changed.  The inode change time of each file is set
9837to the current time.  For example, this code has the same effect as the
9838Unix L<touch(1)> command when the files I<already exist> and belong to
9839the user running the program:
9840
9841    #!/usr/bin/perl
9842    my $atime = my $mtime = time;
9843    utime $atime, $mtime, @ARGV;
9844
9845Since Perl 5.8.0, if the first two elements of the list are
9846L<C<undef>|/undef EXPR>,
9847the L<utime(2)> syscall from your C library is called with a null second
9848argument.  On most systems, this will set the file's access and
9849modification times to the current time (i.e., equivalent to the example
9850above) and will work even on files you don't own provided you have write
9851permission:
9852
9853    for my $file (@ARGV) {
9854	utime(undef, undef, $file)
9855	    || warn "Couldn't touch $file: $!";
9856    }
9857
9858Under NFS this will use the time of the NFS server, not the time of
9859the local machine.  If there is a time synchronization problem, the
9860NFS server and local machine will have different times.  The Unix
9861L<touch(1)> command will in fact normally use this form instead of the
9862one shown in the first example.
9863
9864Passing only one of the first two elements as L<C<undef>|/undef EXPR> is
9865equivalent to passing a 0 and will not have the effect described when
9866both are L<C<undef>|/undef EXPR>.  This also triggers an
9867uninitialized warning.
9868
9869On systems that support L<futimes(2)>, you may pass filehandles among the
9870files.  On systems that don't support L<futimes(2)>, passing filehandles raises
9871an exception.  Filehandles must be passed as globs or glob references to be
9872recognized; barewords are considered filenames.
9873
9874Portability issues: L<perlport/utime>.
9875
9876=item values HASH
9877X<values>
9878
9879=item values ARRAY
9880
9881=for Pod::Functions return a list of the values in a hash
9882
9883In list context, returns a list consisting of all the values of the named
9884hash.  In Perl 5.12 or later only, will also return a list of the values of
9885an array; prior to that release, attempting to use an array argument will
9886produce a syntax error.  In scalar context, returns the number of values.
9887
9888Hash entries are returned in an apparently random order.  The actual random
9889order is specific to a given hash; the exact same series of operations
9890on two hashes may result in a different order for each hash.  Any insertion
9891into the hash may change the order, as will any deletion, with the exception
9892that the most recent key returned by L<C<each>|/each HASH> or
9893L<C<keys>|/keys HASH> may be deleted without changing the order.  So
9894long as a given hash is unmodified you may rely on
9895L<C<keys>|/keys HASH>, L<C<values>|/values HASH> and
9896L<C<each>|/each HASH> to repeatedly return the same order
9897as each other.  See L<perlsec/"Algorithmic Complexity Attacks"> for
9898details on why hash order is randomized.  Aside from the guarantees
9899provided here the exact details of Perl's hash algorithm and the hash
9900traversal order are subject to change in any release of Perl.  Tied hashes
9901may behave differently to Perl's hashes with respect to changes in order on
9902insertion and deletion of items.
9903
9904As a side effect, calling L<C<values>|/values HASH> resets the HASH or
9905ARRAY's internal iterator (see L<C<each>|/each HASH>) before yielding the
9906values.  In particular,
9907calling L<C<values>|/values HASH> in void context resets the iterator
9908with no other overhead.
9909
9910Apart from resetting the iterator,
9911C<values @array> in list context is the same as plain C<@array>.
9912(We recommend that you use void context C<keys @array> for this, but
9913reasoned that taking C<values @array> out would require more
9914documentation than leaving it in.)
9915
9916Note that the values are not copied, which means modifying them will
9917modify the contents of the hash:
9918
9919    for (values %hash)      { s/foo/bar/g }  # modifies %hash values
9920    for (@hash{keys %hash}) { s/foo/bar/g }  # same
9921
9922Starting with Perl 5.14, an experimental feature allowed
9923L<C<values>|/values HASH> to take a
9924scalar expression. This experiment has been deemed unsuccessful, and was
9925removed as of Perl 5.24.
9926
9927To avoid confusing would-be users of your code who are running earlier
9928versions of Perl with mysterious syntax errors, put this sort of thing at
9929the top of your file to signal that your code will work I<only> on Perls of
9930a recent vintage:
9931
9932    use v5.12;	# so keys/values/each work on arrays
9933
9934See also L<C<keys>|/keys HASH>, L<C<each>|/each HASH>, and
9935L<C<sort>|/sort SUBNAME LIST>.
9936
9937=item vec EXPR,OFFSET,BITS
9938X<vec> X<bit> X<bit vector>
9939
9940=for Pod::Functions test or set particular bits in a string
9941
9942Treats the string in EXPR as a bit vector made up of elements of
9943width BITS and returns the value of the element specified by OFFSET
9944as an unsigned integer.  BITS therefore specifies the number of bits
9945that are reserved for each element in the bit vector.  This must
9946be a power of two from 1 to 32 (or 64, if your platform supports
9947that).
9948
9949If BITS is 8, "elements" coincide with bytes of the input string.
9950
9951If BITS is 16 or more, bytes of the input string are grouped into chunks
9952of size BITS/8, and each group is converted to a number as with
9953L<C<pack>|/pack TEMPLATE,LIST>/L<C<unpack>|/unpack TEMPLATE,EXPR> with
9954big-endian formats C<n>/C<N> (and analogously for BITS==64).  See
9955L<C<pack>|/pack TEMPLATE,LIST> for details.
9956
9957If bits is 4 or less, the string is broken into bytes, then the bits
9958of each byte are broken into 8/BITS groups.  Bits of a byte are
9959numbered in a little-endian-ish way, as in C<0x01>, C<0x02>,
9960C<0x04>, C<0x08>, C<0x10>, C<0x20>, C<0x40>, C<0x80>.  For example,
9961breaking the single input byte C<chr(0x36)> into two groups gives a list
9962C<(0x6, 0x3)>; breaking it into 4 groups gives C<(0x2, 0x1, 0x3, 0x0)>.
9963
9964L<C<vec>|/vec EXPR,OFFSET,BITS> may also be assigned to, in which case
9965parentheses are needed
9966to give the expression the correct precedence as in
9967
9968    vec($image, $max_x * $x + $y, 8) = 3;
9969
9970If the selected element is outside the string, the value 0 is returned.
9971If an element off the end of the string is written to, Perl will first
9972extend the string with sufficiently many zero bytes.   It is an error
9973to try to write off the beginning of the string (i.e., negative OFFSET).
9974
9975If the string happens to be encoded as UTF-8 internally (and thus has
9976the UTF8 flag set), L<C<vec>|/vec EXPR,OFFSET,BITS> tries to convert it
9977to use a one-byte-per-character internal representation. However, if the
9978string contains characters with values of 256 or higher, a fatal error
9979will occur.
9980
9981Strings created with L<C<vec>|/vec EXPR,OFFSET,BITS> can also be
9982manipulated with the logical
9983operators C<|>, C<&>, C<^>, and C<~>.  These operators will assume a bit
9984vector operation is desired when both operands are strings.
9985See L<perlop/"Bitwise String Operators">.
9986
9987The following code will build up an ASCII string saying C<'PerlPerlPerl'>.
9988The comments show the string after each step.  Note that this code works
9989in the same way on big-endian or little-endian machines.
9990
9991    my $foo = '';
9992    vec($foo,  0, 32) = 0x5065726C; # 'Perl'
9993
9994    # $foo eq "Perl" eq "\x50\x65\x72\x6C", 32 bits
9995    print vec($foo, 0, 8);  # prints 80 == 0x50 == ord('P')
9996
9997    vec($foo,  2, 16) = 0x5065; # 'PerlPe'
9998    vec($foo,  3, 16) = 0x726C; # 'PerlPerl'
9999    vec($foo,  8,  8) = 0x50;   # 'PerlPerlP'
10000    vec($foo,  9,  8) = 0x65;   # 'PerlPerlPe'
10001    vec($foo, 20,  4) = 2;      # 'PerlPerlPe'   . "\x02"
10002    vec($foo, 21,  4) = 7;      # 'PerlPerlPer'
10003                                   # 'r' is "\x72"
10004    vec($foo, 45,  2) = 3;      # 'PerlPerlPer'  . "\x0c"
10005    vec($foo, 93,  1) = 1;      # 'PerlPerlPer'  . "\x2c"
10006    vec($foo, 94,  1) = 1;      # 'PerlPerlPerl'
10007                                   # 'l' is "\x6c"
10008
10009To transform a bit vector into a string or list of 0's and 1's, use these:
10010
10011    my $bits = unpack("b*", $vector);
10012    my @bits = split(//, unpack("b*", $vector));
10013
10014If you know the exact length in bits, it can be used in place of the C<*>.
10015
10016Here is an example to illustrate how the bits actually fall in place:
10017
10018  #!/usr/bin/perl -wl
10019
10020  print <<'EOT';
10021                                    0         1         2         3
10022                     unpack("V",$_) 01234567890123456789012345678901
10023  ------------------------------------------------------------------
10024  EOT
10025
10026  for $w (0..3) {
10027      $width = 2**$w;
10028      for ($shift=0; $shift < $width; ++$shift) {
10029          for ($off=0; $off < 32/$width; ++$off) {
10030              $str = pack("B*", "0"x32);
10031              $bits = (1<<$shift);
10032              vec($str, $off, $width) = $bits;
10033              $res = unpack("b*",$str);
10034              $val = unpack("V", $str);
10035              write;
10036          }
10037      }
10038  }
10039
10040  format STDOUT =
10041  vec($_,@#,@#) = @<< == @######### @>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
10042  $off, $width, $bits, $val, $res
10043  .
10044  __END__
10045
10046Regardless of the machine architecture on which it runs, the
10047example above should print the following table:
10048
10049                                    0         1         2         3
10050                     unpack("V",$_) 01234567890123456789012345678901
10051  ------------------------------------------------------------------
10052  vec($_, 0, 1) = 1   ==          1 10000000000000000000000000000000
10053  vec($_, 1, 1) = 1   ==          2 01000000000000000000000000000000
10054  vec($_, 2, 1) = 1   ==          4 00100000000000000000000000000000
10055  vec($_, 3, 1) = 1   ==          8 00010000000000000000000000000000
10056  vec($_, 4, 1) = 1   ==         16 00001000000000000000000000000000
10057  vec($_, 5, 1) = 1   ==         32 00000100000000000000000000000000
10058  vec($_, 6, 1) = 1   ==         64 00000010000000000000000000000000
10059  vec($_, 7, 1) = 1   ==        128 00000001000000000000000000000000
10060  vec($_, 8, 1) = 1   ==        256 00000000100000000000000000000000
10061  vec($_, 9, 1) = 1   ==        512 00000000010000000000000000000000
10062  vec($_,10, 1) = 1   ==       1024 00000000001000000000000000000000
10063  vec($_,11, 1) = 1   ==       2048 00000000000100000000000000000000
10064  vec($_,12, 1) = 1   ==       4096 00000000000010000000000000000000
10065  vec($_,13, 1) = 1   ==       8192 00000000000001000000000000000000
10066  vec($_,14, 1) = 1   ==      16384 00000000000000100000000000000000
10067  vec($_,15, 1) = 1   ==      32768 00000000000000010000000000000000
10068  vec($_,16, 1) = 1   ==      65536 00000000000000001000000000000000
10069  vec($_,17, 1) = 1   ==     131072 00000000000000000100000000000000
10070  vec($_,18, 1) = 1   ==     262144 00000000000000000010000000000000
10071  vec($_,19, 1) = 1   ==     524288 00000000000000000001000000000000
10072  vec($_,20, 1) = 1   ==    1048576 00000000000000000000100000000000
10073  vec($_,21, 1) = 1   ==    2097152 00000000000000000000010000000000
10074  vec($_,22, 1) = 1   ==    4194304 00000000000000000000001000000000
10075  vec($_,23, 1) = 1   ==    8388608 00000000000000000000000100000000
10076  vec($_,24, 1) = 1   ==   16777216 00000000000000000000000010000000
10077  vec($_,25, 1) = 1   ==   33554432 00000000000000000000000001000000
10078  vec($_,26, 1) = 1   ==   67108864 00000000000000000000000000100000
10079  vec($_,27, 1) = 1   ==  134217728 00000000000000000000000000010000
10080  vec($_,28, 1) = 1   ==  268435456 00000000000000000000000000001000
10081  vec($_,29, 1) = 1   ==  536870912 00000000000000000000000000000100
10082  vec($_,30, 1) = 1   == 1073741824 00000000000000000000000000000010
10083  vec($_,31, 1) = 1   == 2147483648 00000000000000000000000000000001
10084  vec($_, 0, 2) = 1   ==          1 10000000000000000000000000000000
10085  vec($_, 1, 2) = 1   ==          4 00100000000000000000000000000000
10086  vec($_, 2, 2) = 1   ==         16 00001000000000000000000000000000
10087  vec($_, 3, 2) = 1   ==         64 00000010000000000000000000000000
10088  vec($_, 4, 2) = 1   ==        256 00000000100000000000000000000000
10089  vec($_, 5, 2) = 1   ==       1024 00000000001000000000000000000000
10090  vec($_, 6, 2) = 1   ==       4096 00000000000010000000000000000000
10091  vec($_, 7, 2) = 1   ==      16384 00000000000000100000000000000000
10092  vec($_, 8, 2) = 1   ==      65536 00000000000000001000000000000000
10093  vec($_, 9, 2) = 1   ==     262144 00000000000000000010000000000000
10094  vec($_,10, 2) = 1   ==    1048576 00000000000000000000100000000000
10095  vec($_,11, 2) = 1   ==    4194304 00000000000000000000001000000000
10096  vec($_,12, 2) = 1   ==   16777216 00000000000000000000000010000000
10097  vec($_,13, 2) = 1   ==   67108864 00000000000000000000000000100000
10098  vec($_,14, 2) = 1   ==  268435456 00000000000000000000000000001000
10099  vec($_,15, 2) = 1   == 1073741824 00000000000000000000000000000010
10100  vec($_, 0, 2) = 2   ==          2 01000000000000000000000000000000
10101  vec($_, 1, 2) = 2   ==          8 00010000000000000000000000000000
10102  vec($_, 2, 2) = 2   ==         32 00000100000000000000000000000000
10103  vec($_, 3, 2) = 2   ==        128 00000001000000000000000000000000
10104  vec($_, 4, 2) = 2   ==        512 00000000010000000000000000000000
10105  vec($_, 5, 2) = 2   ==       2048 00000000000100000000000000000000
10106  vec($_, 6, 2) = 2   ==       8192 00000000000001000000000000000000
10107  vec($_, 7, 2) = 2   ==      32768 00000000000000010000000000000000
10108  vec($_, 8, 2) = 2   ==     131072 00000000000000000100000000000000
10109  vec($_, 9, 2) = 2   ==     524288 00000000000000000001000000000000
10110  vec($_,10, 2) = 2   ==    2097152 00000000000000000000010000000000
10111  vec($_,11, 2) = 2   ==    8388608 00000000000000000000000100000000
10112  vec($_,12, 2) = 2   ==   33554432 00000000000000000000000001000000
10113  vec($_,13, 2) = 2   ==  134217728 00000000000000000000000000010000
10114  vec($_,14, 2) = 2   ==  536870912 00000000000000000000000000000100
10115  vec($_,15, 2) = 2   == 2147483648 00000000000000000000000000000001
10116  vec($_, 0, 4) = 1   ==          1 10000000000000000000000000000000
10117  vec($_, 1, 4) = 1   ==         16 00001000000000000000000000000000
10118  vec($_, 2, 4) = 1   ==        256 00000000100000000000000000000000
10119  vec($_, 3, 4) = 1   ==       4096 00000000000010000000000000000000
10120  vec($_, 4, 4) = 1   ==      65536 00000000000000001000000000000000
10121  vec($_, 5, 4) = 1   ==    1048576 00000000000000000000100000000000
10122  vec($_, 6, 4) = 1   ==   16777216 00000000000000000000000010000000
10123  vec($_, 7, 4) = 1   ==  268435456 00000000000000000000000000001000
10124  vec($_, 0, 4) = 2   ==          2 01000000000000000000000000000000
10125  vec($_, 1, 4) = 2   ==         32 00000100000000000000000000000000
10126  vec($_, 2, 4) = 2   ==        512 00000000010000000000000000000000
10127  vec($_, 3, 4) = 2   ==       8192 00000000000001000000000000000000
10128  vec($_, 4, 4) = 2   ==     131072 00000000000000000100000000000000
10129  vec($_, 5, 4) = 2   ==    2097152 00000000000000000000010000000000
10130  vec($_, 6, 4) = 2   ==   33554432 00000000000000000000000001000000
10131  vec($_, 7, 4) = 2   ==  536870912 00000000000000000000000000000100
10132  vec($_, 0, 4) = 4   ==          4 00100000000000000000000000000000
10133  vec($_, 1, 4) = 4   ==         64 00000010000000000000000000000000
10134  vec($_, 2, 4) = 4   ==       1024 00000000001000000000000000000000
10135  vec($_, 3, 4) = 4   ==      16384 00000000000000100000000000000000
10136  vec($_, 4, 4) = 4   ==     262144 00000000000000000010000000000000
10137  vec($_, 5, 4) = 4   ==    4194304 00000000000000000000001000000000
10138  vec($_, 6, 4) = 4   ==   67108864 00000000000000000000000000100000
10139  vec($_, 7, 4) = 4   == 1073741824 00000000000000000000000000000010
10140  vec($_, 0, 4) = 8   ==          8 00010000000000000000000000000000
10141  vec($_, 1, 4) = 8   ==        128 00000001000000000000000000000000
10142  vec($_, 2, 4) = 8   ==       2048 00000000000100000000000000000000
10143  vec($_, 3, 4) = 8   ==      32768 00000000000000010000000000000000
10144  vec($_, 4, 4) = 8   ==     524288 00000000000000000001000000000000
10145  vec($_, 5, 4) = 8   ==    8388608 00000000000000000000000100000000
10146  vec($_, 6, 4) = 8   ==  134217728 00000000000000000000000000010000
10147  vec($_, 7, 4) = 8   == 2147483648 00000000000000000000000000000001
10148  vec($_, 0, 8) = 1   ==          1 10000000000000000000000000000000
10149  vec($_, 1, 8) = 1   ==        256 00000000100000000000000000000000
10150  vec($_, 2, 8) = 1   ==      65536 00000000000000001000000000000000
10151  vec($_, 3, 8) = 1   ==   16777216 00000000000000000000000010000000
10152  vec($_, 0, 8) = 2   ==          2 01000000000000000000000000000000
10153  vec($_, 1, 8) = 2   ==        512 00000000010000000000000000000000
10154  vec($_, 2, 8) = 2   ==     131072 00000000000000000100000000000000
10155  vec($_, 3, 8) = 2   ==   33554432 00000000000000000000000001000000
10156  vec($_, 0, 8) = 4   ==          4 00100000000000000000000000000000
10157  vec($_, 1, 8) = 4   ==       1024 00000000001000000000000000000000
10158  vec($_, 2, 8) = 4   ==     262144 00000000000000000010000000000000
10159  vec($_, 3, 8) = 4   ==   67108864 00000000000000000000000000100000
10160  vec($_, 0, 8) = 8   ==          8 00010000000000000000000000000000
10161  vec($_, 1, 8) = 8   ==       2048 00000000000100000000000000000000
10162  vec($_, 2, 8) = 8   ==     524288 00000000000000000001000000000000
10163  vec($_, 3, 8) = 8   ==  134217728 00000000000000000000000000010000
10164  vec($_, 0, 8) = 16  ==         16 00001000000000000000000000000000
10165  vec($_, 1, 8) = 16  ==       4096 00000000000010000000000000000000
10166  vec($_, 2, 8) = 16  ==    1048576 00000000000000000000100000000000
10167  vec($_, 3, 8) = 16  ==  268435456 00000000000000000000000000001000
10168  vec($_, 0, 8) = 32  ==         32 00000100000000000000000000000000
10169  vec($_, 1, 8) = 32  ==       8192 00000000000001000000000000000000
10170  vec($_, 2, 8) = 32  ==    2097152 00000000000000000000010000000000
10171  vec($_, 3, 8) = 32  ==  536870912 00000000000000000000000000000100
10172  vec($_, 0, 8) = 64  ==         64 00000010000000000000000000000000
10173  vec($_, 1, 8) = 64  ==      16384 00000000000000100000000000000000
10174  vec($_, 2, 8) = 64  ==    4194304 00000000000000000000001000000000
10175  vec($_, 3, 8) = 64  == 1073741824 00000000000000000000000000000010
10176  vec($_, 0, 8) = 128 ==        128 00000001000000000000000000000000
10177  vec($_, 1, 8) = 128 ==      32768 00000000000000010000000000000000
10178  vec($_, 2, 8) = 128 ==    8388608 00000000000000000000000100000000
10179  vec($_, 3, 8) = 128 == 2147483648 00000000000000000000000000000001
10180
10181=item wait
10182X<wait>
10183
10184=for Pod::Functions wait for any child process to die
10185
10186Behaves like L<wait(2)> on your system: it waits for a child
10187process to terminate and returns the pid of the deceased process, or
10188C<-1> if there are no child processes.  The status is returned in
10189L<C<$?>|perlvar/$?> and
10190L<C<${^CHILD_ERROR_NATIVE}>|perlvar/${^CHILD_ERROR_NATIVE}>.
10191Note that a return value of C<-1> could mean that child processes are
10192being automatically reaped, as described in L<perlipc>.
10193
10194If you use L<C<wait>|/wait> in your handler for
10195L<C<$SIG{CHLD}>|perlvar/%SIG>, it may accidentally wait for the child
10196created by L<C<qx>|/qxE<sol>STRINGE<sol>> or L<C<system>|/system LIST>.
10197See L<perlipc> for details.
10198
10199Portability issues: L<perlport/wait>.
10200
10201=item waitpid PID,FLAGS
10202X<waitpid>
10203
10204=for Pod::Functions wait for a particular child process to die
10205
10206Waits for a particular child process to terminate and returns the pid of
10207the deceased process, or C<-1> if there is no such child process.  A
10208non-blocking wait (with L<WNOHANG|POSIX/C<WNOHANG>> in FLAGS) can return 0 if
10209there are child processes matching PID but none have terminated yet.
10210The status is returned in L<C<$?>|perlvar/$?> and
10211L<C<${^CHILD_ERROR_NATIVE}>|perlvar/${^CHILD_ERROR_NATIVE}>.
10212
10213A PID of C<0> indicates to wait for any child process whose process group ID is
10214equal to that of the current process.  A PID of less than C<-1> indicates to
10215wait for any child process whose process group ID is equal to -PID.  A PID of
10216C<-1> indicates to wait for any child process.
10217
10218If you say
10219
10220    use POSIX ":sys_wait_h";
10221
10222    my $kid;
10223    do {
10224        $kid = waitpid(-1, WNOHANG);
10225    } while $kid > 0;
10226
10227or
10228
10229    1 while waitpid(-1, WNOHANG) > 0;
10230
10231then you can do a non-blocking wait for all pending zombie processes (see
10232L<POSIX/WAIT>).
10233Non-blocking wait is available on machines supporting either the
10234L<waitpid(2)> or L<wait4(2)> syscalls.  However, waiting for a particular
10235pid with FLAGS of C<0> is implemented everywhere.  (Perl emulates the
10236system call by remembering the status values of processes that have
10237exited but have not been harvested by the Perl script yet.)
10238
10239Note that on some systems, a return value of C<-1> could mean that child
10240processes are being automatically reaped.  See L<perlipc> for details,
10241and for other examples.
10242
10243Portability issues: L<perlport/waitpid>.
10244
10245=item wantarray
10246X<wantarray> X<context>
10247
10248=for Pod::Functions get void vs scalar vs list context of current subroutine call
10249
10250Returns true if the context of the currently executing subroutine or
10251L<C<eval>|/eval EXPR> is looking for a list value.  Returns false if the
10252context is
10253looking for a scalar.  Returns the undefined value if the context is
10254looking for no value (void context).
10255
10256    return unless defined wantarray; # don't bother doing more
10257    my @a = complex_calculation();
10258    return wantarray ? @a : "@a";
10259
10260L<C<wantarray>|/wantarray>'s result is unspecified in the top level of a file,
10261in a C<BEGIN>, C<UNITCHECK>, C<CHECK>, C<INIT> or C<END> block, or
10262in a C<DESTROY> method.
10263
10264This function should have been named wantlist() instead.
10265
10266=item warn LIST
10267X<warn> X<warning> X<STDERR>
10268
10269=for Pod::Functions print debugging info
10270
10271Emits a warning, usually by printing it to C<STDERR>.  C<warn> interprets
10272its operand LIST in the same way as C<die>, but is slightly different
10273in what it defaults to when LIST is empty or makes an empty string.
10274If it is empty and L<C<$@>|perlvar/$@> already contains an exception
10275value then that value is used after appending C<"\t...caught">.  If it
10276is empty and C<$@> is also empty then the string C<"Warning: Something's
10277wrong"> is used.
10278
10279By default, the exception derived from the operand LIST is stringified
10280and printed to C<STDERR>.  This behaviour can be altered by installing
10281a L<C<$SIG{__WARN__}>|perlvar/%SIG> handler.  If there is such a
10282handler then no message is automatically printed; it is the handler's
10283responsibility to deal with the exception
10284as it sees fit (like, for instance, converting it into a
10285L<C<die>|/die LIST>).  Most
10286handlers must therefore arrange to actually display the
10287warnings that they are not prepared to deal with, by calling
10288L<C<warn>|/warn LIST>
10289again in the handler.  Note that this is quite safe and will not
10290produce an endless loop, since C<__WARN__> hooks are not called from
10291inside one.
10292
10293You will find this behavior is slightly different from that of
10294L<C<$SIG{__DIE__}>|perlvar/%SIG> handlers (which don't suppress the
10295error text, but can instead call L<C<die>|/die LIST> again to change
10296it).
10297
10298Using a C<__WARN__> handler provides a powerful way to silence all
10299warnings (even the so-called mandatory ones).  An example:
10300
10301    # wipe out *all* compile-time warnings
10302    BEGIN { $SIG{'__WARN__'} = sub { warn $_[0] if $DOWARN } }
10303    my $foo = 10;
10304    my $foo = 20;          # no warning about duplicate my $foo,
10305                           # but hey, you asked for it!
10306    # no compile-time or run-time warnings before here
10307    $DOWARN = 1;
10308
10309    # run-time warnings enabled after here
10310    warn "\$foo is alive and $foo!";     # does show up
10311
10312See L<perlvar> for details on setting L<C<%SIG>|perlvar/%SIG> entries
10313and for more
10314examples.  See the L<Carp> module for other kinds of warnings using its
10315C<carp> and C<cluck> functions.
10316
10317=item write FILEHANDLE
10318X<write>
10319
10320=item write EXPR
10321
10322=item write
10323
10324=for Pod::Functions print a picture record
10325
10326Writes a formatted record (possibly multi-line) to the specified FILEHANDLE,
10327using the format associated with that file.  By default the format for
10328a file is the one having the same name as the filehandle, but the
10329format for the current output channel (see the
10330L<C<select>|/select FILEHANDLE> function) may be set explicitly by
10331assigning the name of the format to the L<C<$~>|perlvar/$~> variable.
10332
10333Top of form processing is handled automatically:  if there is insufficient
10334room on the current page for the formatted record, the page is advanced by
10335writing a form feed and a special top-of-page
10336format is used to format the new
10337page header before the record is written.  By default, the top-of-page
10338format is the name of the filehandle with C<_TOP> appended, or C<top>
10339in the current package if the former does not exist.  This would be a
10340problem with autovivified filehandles, but it may be dynamically set to the
10341format of your choice by assigning the name to the L<C<$^>|perlvar/$^>
10342variable while that filehandle is selected.  The number of lines
10343remaining on the current page is in variable L<C<$->|perlvar/$->, which
10344can be set to C<0> to force a new page.
10345
10346If FILEHANDLE is unspecified, output goes to the current default output
10347channel, which starts out as STDOUT but may be changed by the
10348L<C<select>|/select FILEHANDLE> operator.  If the FILEHANDLE is an EXPR,
10349then the expression
10350is evaluated and the resulting string is used to look up the name of
10351the FILEHANDLE at run time.  For more on formats, see L<perlform>.
10352
10353Note that write is I<not> the opposite of
10354L<C<read>|/read FILEHANDLE,SCALAR,LENGTH,OFFSET>.  Unfortunately.
10355
10356=item y///
10357
10358=for Pod::Functions transliterate a string
10359
10360The transliteration operator.  Same as
10361L<C<trE<sol>E<sol>E<sol>>|/trE<sol>E<sol>E<sol>>.  See
10362L<perlop/"Quote-Like Operators">.
10363
10364=back
10365
10366=head2 Non-function Keywords by Cross-reference
10367
10368=head3 perldata
10369
10370=over
10371
10372=item __DATA__
10373
10374=item __END__
10375
10376These keywords are documented in L<perldata/"Special Literals">.
10377
10378=back
10379
10380=head3 perlmod
10381
10382=over
10383
10384=item BEGIN
10385
10386=item CHECK
10387
10388=item END
10389
10390=item INIT
10391
10392=item UNITCHECK
10393
10394These compile phase keywords are documented in L<perlmod/"BEGIN, UNITCHECK, CHECK, INIT and END">.
10395
10396=back
10397
10398=head3 perlobj
10399
10400=over
10401
10402=item DESTROY
10403
10404This method keyword is documented in L<perlobj/"Destructors">.
10405
10406=back
10407
10408=head3 perlop
10409
10410=over
10411
10412=item and
10413
10414=item cmp
10415
10416=item eq
10417
10418=item ge
10419
10420=item gt
10421
10422=item isa
10423
10424=item le
10425
10426=item lt
10427
10428=item ne
10429
10430=item not
10431
10432=item or
10433
10434=item x
10435
10436=item xor
10437
10438These operators are documented in L<perlop>.
10439
10440=back
10441
10442=head3 perlsub
10443
10444=over
10445
10446=item AUTOLOAD
10447
10448This keyword is documented in L<perlsub/"Autoloading">.
10449
10450=back
10451
10452=head3 perlsyn
10453
10454=over
10455
10456=item else
10457
10458=item elsif
10459
10460=item for
10461
10462=item foreach
10463
10464=item if
10465
10466=item unless
10467
10468=item until
10469
10470=item while
10471
10472These flow-control keywords are documented in L<perlsyn/"Compound Statements">.
10473
10474=item elseif
10475
10476The "else if" keyword is spelled C<elsif> in Perl.  There's no C<elif>
10477or C<else if> either.  It does parse C<elseif>, but only to warn you
10478about not using it.
10479
10480See the documentation for flow-control keywords in L<perlsyn/"Compound
10481Statements">.
10482
10483=back
10484
10485=over
10486
10487=item default
10488
10489=item given
10490
10491=item when
10492
10493These flow-control keywords related to the experimental switch feature are
10494documented in L<perlsyn/"Switch Statements">.
10495
10496=back
10497
10498=over
10499
10500=item try
10501
10502=item catch
10503
10504=item finally
10505
10506These flow-control keywords related to the experimental C<try> feature are
10507documented in L<perlsyn/"Try Catch Exception Handling">.
10508
10509=back
10510
10511=over
10512
10513=item defer
10514
10515This flow-control keyword related to the experimental C<defer> feature is
10516documented in L<perlsyn/"defer blocks">.
10517
10518=back
10519
10520=cut
10521