3 perlrun - how to execute the Perl interpreter
7 B<perl> S<[ B<-sTtuUWX> ]>
8 S<[ B<-hv> ] [ B<-V>[:I<configvar>] ]>
9 S<[ B<-cw> ] [ B<-d>[B<t>][:I<debugger>] ] [ B<-D>[I<number/list>] ]>
10 S<[ B<-pna> ] [ B<-F>I<pattern> ] [ B<-l>[I<octal>] ] [ B<-0>[I<octal/hexadecimal>] ]>
11 S<[ B<-I>I<dir> ] [ B<-m>[B<->]I<module> ] [ B<-M>[B<->]I<'module...'> ] [ B<-f> ]>
12 S<[ B<-C [I<number/list>] >]>
15 S<[ B<-i>[I<extension>] ]>
16 S<[ [B<-e>|B<-E>] I<'command'> ] [ B<--> ] [ I<programfile> ] [ I<argument> ]...>
20 The normal way to run a Perl program is by making it directly
21 executable, or else by passing the name of the source file as an
22 argument on the command line. (An interactive Perl environment
23 is also possible--see L<perldebug> for details on how to do that.)
24 Upon startup, Perl looks for your program in one of the following
31 Specified line by line via B<-e> or B<-E> switches on the command line.
35 Contained in the file specified by the first filename on the command line.
36 (Note that systems supporting the #! notation invoke interpreters this
37 way. See L<Location of Perl>.)
41 Passed in implicitly via standard input. This works only if there are
42 no filename arguments--to pass arguments to a STDIN-read program you
43 must explicitly specify a "-" for the program name.
47 With methods 2 and 3, Perl starts parsing the input file from the
48 beginning, unless you've specified a B<-x> switch, in which case it
49 scans for the first line starting with #! and containing the word
50 "perl", and starts there instead. This is useful for running a program
51 embedded in a larger message. (In this case you would indicate the end
52 of the program using the C<__END__> token.)
54 The #! line is always examined for switches as the line is being
55 parsed. Thus, if you're on a machine that allows only one argument
56 with the #! line, or worse, doesn't even recognize the #! line, you
57 still can get consistent switch behavior regardless of how Perl was
58 invoked, even if B<-x> was used to find the beginning of the program.
60 Because historically some operating systems silently chopped off
61 kernel interpretation of the #! line after 32 characters, some
62 switches may be passed in on the command line, and some may not;
63 you could even get a "-" without its letter, if you're not careful.
64 You probably want to make sure that all your switches fall either
65 before or after that 32-character boundary. Most switches don't
66 actually care if they're processed redundantly, but getting a "-"
67 instead of a complete switch could cause Perl to try to execute
68 standard input instead of your program. And a partial B<-I> switch
69 could also cause odd results.
71 Some switches do care if they are processed twice, for instance
72 combinations of B<-l> and B<-0>. Either put all the switches after
73 the 32-character boundary (if applicable), or replace the use of
74 B<-0>I<digits> by C<BEGIN{ $/ = "\0digits"; }>.
76 Parsing of the #! switches starts wherever "perl" is mentioned in the line.
77 The sequences "-*" and "- " are specifically ignored so that you could,
78 if you were so inclined, say
82 eval 'exec perl -x -wS $0 ${1+"$@"}'
85 to let Perl see the B<-p> switch.
87 A similar trick involves the B<env> program, if you have it.
91 The examples above use a relative path to the perl interpreter,
92 getting whatever version is first in the user's path. If you want
93 a specific version of Perl, say, perl5.005_57, you should place
94 that directly in the #! line's path.
96 If the #! line does not contain the word "perl", the program named after
97 the #! is executed instead of the Perl interpreter. This is slightly
98 bizarre, but it helps people on machines that don't do #!, because they
99 can tell a program that their SHELL is F</usr/bin/perl>, and Perl will then
100 dispatch the program to the correct interpreter for them.
102 After locating your program, Perl compiles the entire program to an
103 internal form. If there are any compilation errors, execution of the
104 program is not attempted. (This is unlike the typical shell script,
105 which might run part-way through before finding a syntax error.)
107 If the program is syntactically correct, it is executed. If the program
108 runs off the end without hitting an exit() or die() operator, an implicit
109 C<exit(0)> is provided to indicate successful completion.
111 =head2 #! and quoting on non-Unix systems
114 Unix's #! technique can be simulated on other systems:
122 extproc perl -S -your_switches
124 as the first line in C<*.cmd> file (B<-S> due to a bug in cmd.exe's
129 Create a batch file to run your program, and codify it in
130 C<ALTERNATE_SHEBANG> (see the F<dosish.h> file in the source
131 distribution for more information).
135 The Win95/NT installation, when using the ActiveState installer for Perl,
136 will modify the Registry to associate the F<.pl> extension with the perl
137 interpreter. If you install Perl by other means (including building from
138 the sources), you may have to modify the Registry yourself. Note that
139 this means you can no longer tell the difference between an executable
140 Perl program and a Perl library file.
146 $ perl -mysw 'f$env("procedure")' 'p1' 'p2' 'p3' 'p4' 'p5' 'p6' 'p7' 'p8' !
147 $ exit++ + ++$status != 0 and $exit = $status = undef;
149 at the top of your program, where B<-mysw> are any command line switches you
150 want to pass to Perl. You can now invoke the program directly, by saying
151 C<perl program>, or as a DCL procedure, by saying C<@program> (or implicitly
152 via F<DCL$PATH> by just using the name of the program).
154 This incantation is a bit much to remember, but Perl will display it for
155 you if you say C<perl "-V:startperl">.
159 Command-interpreters on non-Unix systems have rather different ideas
160 on quoting than Unix shells. You'll need to learn the special
161 characters in your command-interpreter (C<*>, C<\> and C<"> are
162 common) and how to protect whitespace and these characters to run
163 one-liners (see B<-e> below).
165 On some systems, you may have to change single-quotes to double ones,
166 which you must I<not> do on Unix or Plan 9 systems. You might also
167 have to change a single % to a %%.
172 perl -e 'print "Hello world\n"'
175 perl -e "print \"Hello world\n\""
178 perl -e "print ""Hello world\n"""
180 The problem is that none of this is reliable: it depends on the
181 command and it is entirely possible neither works. If B<4DOS> were
182 the command shell, this would probably work better:
184 perl -e "print <Ctrl-x>"Hello world\n<Ctrl-x>""
186 B<CMD.EXE> in Windows NT slipped a lot of standard Unix functionality in
187 when nobody was looking, but just try to find documentation for its
190 There is no general solution to all of this. It's just a mess.
192 =head2 Location of Perl
193 X<perl, location of interpreter>
195 It may seem obvious to say, but Perl is useful only when users can
196 easily find it. When possible, it's good for both F</usr/bin/perl>
197 and F</usr/local/bin/perl> to be symlinks to the actual binary. If
198 that can't be done, system administrators are strongly encouraged
199 to put (symlinks to) perl and its accompanying utilities into a
200 directory typically found along a user's PATH, or in some other
201 obvious and convenient place.
203 In this documentation, C<#!/usr/bin/perl> on the first line of the program
204 will stand in for whatever method works on your system. You are
205 advised to use a specific path if you care about a specific version.
207 #!/usr/local/bin/perl5.00554
209 or if you just want to be running at least version, place a statement
210 like this at the top of your program:
214 =head2 Command Switches
215 X<perl, command switches> X<command switches>
217 As with all standard commands, a single-character switch may be
218 clustered with the following switch, if any.
220 #!/usr/bin/perl -spi.orig # same as -s -p -i.orig
226 =item B<-0>[I<octal/hexadecimal>]
229 specifies the input record separator (C<$/>) as an octal or
230 hexadecimal number. If there are no digits, the null character is the
231 separator. Other switches may precede or follow the digits. For
232 example, if you have a version of B<find> which can print filenames
233 terminated by the null character, you can say this:
235 find . -name '*.orig' -print0 | perl -n0e unlink
237 The special value 00 will cause Perl to slurp files in paragraph mode.
238 Any value 0400 or above will cause Perl to slurp files whole, but by convention
239 the value 0777 is the one normally used for this purpose.
241 You can also specify the separator character using hexadecimal notation:
242 C<-0xHHH...>, where the C<H> are valid hexadecimal digits. Unlike the octal
243 form, this one may be used to specify any Unicode character, even those beyond
245 (This means that you cannot use the C<-x> with a directory name that
246 consists of hexadecimal digits.)
251 turns on autosplit mode when used with a B<-n> or B<-p>. An implicit
252 split command to the @F array is done as the first thing inside the
253 implicit while loop produced by the B<-n> or B<-p>.
255 perl -ane 'print pop(@F), "\n";'
264 An alternate delimiter may be specified using B<-F>.
266 =item B<-C [I<number/list>]>
269 The C<-C> flag controls some of the Perl Unicode features.
271 As of 5.8.1, the C<-C> can be followed either by a number or a list
272 of option letters. The letters, their numeric values, and effects
273 are as follows; listing the letters is equal to summing the numbers.
275 I 1 STDIN is assumed to be in UTF-8
276 O 2 STDOUT will be in UTF-8
277 E 4 STDERR will be in UTF-8
279 i 8 UTF-8 is the default PerlIO layer for input streams
280 o 16 UTF-8 is the default PerlIO layer for output streams
282 A 32 the @ARGV elements are expected to be strings encoded
284 L 64 normally the "IOEioA" are unconditional,
285 the L makes them conditional on the locale environment
286 variables (the LC_ALL, LC_TYPE, and LANG, in the order
287 of decreasing precedence) -- if the variables indicate
288 UTF-8, then the selected "IOEioA" are in effect
289 a 256 Set ${^UTF8CACHE} to -1, to run the UTF-8 caching code in
292 =for documenting_the_underdocumented
293 perl.h gives W/128 as PERL_UNICODE_WIDESYSCALLS "/* for Sarathy */"
296 perltodo mentions Unicode in %ENV and filenames. I guess that these will be
297 options e and f (or F).
299 For example, C<-COE> and C<-C6> will both turn on UTF-8-ness on both
300 STDOUT and STDERR. Repeating letters is just redundant, not cumulative
303 The C<io> options mean that any subsequent open() (or similar I/O
304 operations) in the current file scope will have the C<:utf8> PerlIO layer
305 implicitly applied to them, in other words, UTF-8 is expected from any
306 input stream, and UTF-8 is produced to any output stream. This is just
307 the default, with explicit layers in open() and with binmode() one can
308 manipulate streams as usual.
310 C<-C> on its own (not followed by any number or option list), or the
311 empty string C<""> for the C<PERL_UNICODE> environment variable, has the
312 same effect as C<-CSDL>. In other words, the standard I/O handles and
313 the default C<open()> layer are UTF-8-fied B<but> only if the locale
314 environment variables indicate a UTF-8 locale. This behaviour follows
315 the I<implicit> (and problematic) UTF-8 behaviour of Perl 5.8.0.
317 You can use C<-C0> (or C<"0"> for C<PERL_UNICODE>) to explicitly
318 disable all the above Unicode features.
320 The read-only magic variable C<${^UNICODE}> reflects the numeric value
321 of this setting. This variable is set during Perl startup and is
322 thereafter read-only. If you want runtime effects, use the three-arg
323 open() (see L<perlfunc/open>), the two-arg binmode() (see L<perlfunc/binmode>),
324 and the C<open> pragma (see L<open>).
326 (In Perls earlier than 5.8.1 the C<-C> switch was a Win32-only switch
327 that enabled the use of Unicode-aware "wide system call" Win32 APIs.
328 This feature was practically unused, however, and the command line
329 switch was therefore "recycled".)
331 B<Note:> Since perl 5.10.1, if the -C option is used on the #! line, it
332 must be specified on the command line as well, since the standard streams
333 are already set up at this point in the execution of the perl interpreter.
334 You can also use binmode() to set the encoding of an I/O stream.
339 causes Perl to check the syntax of the program and then exit without
340 executing it. Actually, it I<will> execute C<BEGIN>, C<UNITCHECK>,
341 C<CHECK>, and C<use> blocks, because these are considered as occurring
342 outside the execution of your program. C<INIT> and C<END> blocks,
343 however, will be skipped.
350 runs the program under the Perl debugger. See L<perldebug>.
351 If B<t> is specified, it indicates to the debugger that threads
352 will be used in the code being debugged.
354 =item B<-d:>I<foo[=bar,baz]>
357 =item B<-dt:>I<foo[=bar,baz]>
359 runs the program under the control of a debugging, profiling, or
360 tracing module installed as Devel::foo. E.g., B<-d:DProf> executes
361 the program using the Devel::DProf profiler. As with the B<-M>
362 flag, options may be passed to the Devel::foo package where they
363 will be received and interpreted by the Devel::foo::import routine.
364 Again, like B<-M>, use -d:-foo to call Devel::foo::unimport instead of import.
365 The comma-separated list of options must follow a C<=> character.
366 If B<t> is specified, it indicates to the debugger that threads
367 will be used in the code being debugged.
370 =item B<-D>I<letters>
371 X<-D> X<DEBUGGING> X<-DDEBUGGING>
375 sets debugging flags. To watch how it executes your program, use
376 B<-Dtls>. (This works only if debugging is compiled into your
377 Perl.) Another nice value is B<-Dx>, which lists your compiled
378 syntax tree. And B<-Dr> displays compiled regular expressions;
379 the format of the output is explained in L<perldebguts>.
381 As an alternative, specify a number instead of list of letters (e.g.,
382 B<-D14> is equivalent to B<-Dtls>):
384 1 p Tokenizing and parsing (with v, displays parse stack)
385 2 s Stack snapshots (with v, displays all stacks)
386 4 l Context (loop) stack processing
388 16 o Method and overloading resolution
389 32 c String/numeric conversions
390 64 P Print profiling info, source file input state
391 128 m Memory and SV allocation
392 256 f Format processing
393 512 r Regular expression parsing and execution
394 1024 x Syntax tree dump
395 2048 u Tainting checks
396 4096 U Unofficial, User hacking (reserved for private, unreleased use)
397 8192 H Hash dump -- usurps values()
398 16384 X Scratchpad allocation
401 262144 R Include reference counts of dumped variables (eg when using -Ds)
402 524288 J show s,t,P-debug (don't Jump over) on opcodes within package DB
403 1048576 v Verbose: use in conjunction with other flags
404 2097152 C Copy On Write
405 4194304 A Consistency checks on internal structures
406 8388608 q quiet - currently only suppresses the "EXECUTING" message
407 16777216 M trace smart match resolution
408 33554432 B dump suBroutine definitions, including special Blocks like BEGIN
410 All these flags require B<-DDEBUGGING> when you compile the Perl
411 executable (but see C<:opd> in L<Devel::Peek> or L<re/'debug' mode>
412 which may change this).
413 See the F<INSTALL> file in the Perl source distribution
414 for how to do this. This flag is automatically set if you include B<-g>
415 option when C<Configure> asks you about optimizer/debugger flags.
417 If you're just trying to get a print out of each line of Perl code
418 as it executes, the way that C<sh -x> provides for shell scripts,
419 you can't use Perl's B<-D> switch. Instead do this
421 # If you have "env" utility
422 env PERLDB_OPTS="NonStop=1 AutoTrace=1 frame=2" perl -dS program
424 # Bourne shell syntax
425 $ PERLDB_OPTS="NonStop=1 AutoTrace=1 frame=2" perl -dS program
428 % (setenv PERLDB_OPTS "NonStop=1 AutoTrace=1 frame=2"; perl -dS program)
430 See L<perldebug> for details and variations.
432 =item B<-e> I<commandline>
435 may be used to enter one line of program. If B<-e> is given, Perl
436 will not look for a filename in the argument list. Multiple B<-e>
437 commands may be given to build up a multi-line script. Make sure
438 to use semicolons where you would in a normal program.
440 =item B<-E> I<commandline>
443 behaves just like B<-e>, except that it implicitly enables all
444 optional features (in the main compilation unit). See L<feature>.
447 X<-f> X<sitecustomize> X<sitecustomize.pl>
449 Disable executing F<$Config{sitelib}/sitecustomize.pl> at startup.
451 Perl can be built so that it by default will try to execute
452 F<$Config{sitelib}/sitecustomize.pl> at startup (in a BEGIN block).
453 This is a hook that allows the sysadmin to customize how perl behaves.
454 It can for instance be used to add entries to the @INC array to make perl
455 find modules in non-standard locations.
457 Perl actually inserts the following code:
460 do { local $!; -f "$Config{sitelib}/sitecustomize.pl"; }
461 && do "$Config{sitelib}/sitecustomize.pl";
464 Since it is an actual C<do> (not a C<require>), F<sitecustomize.pl>
465 doesn't need to return a true value. The code is run in package C<main>,
466 in its own lexical scope. However, if the script dies, C<$@> will not
469 The value of C<$Config{sitelib}> is also determined in C code and not
470 read from C<Config.pm>, which is not loaded.
472 The code is executed B<very> early. For example, any changes made to
473 C<@INC> will show up in the output of `perl -V`. Of course, C<END>
474 blocks will be likewise executed very late.
476 To determine at runtime if this capability has been compiled in your
477 perl, you can check the value of C<$Config{usesitecustomize}>.
479 =item B<-F>I<pattern>
482 specifies the pattern to split on if B<-a> is also in effect. The
483 pattern may be surrounded by C<//>, C<"">, or C<''>, otherwise it will be
484 put in single quotes. You can't use literal whitespace in the pattern.
489 prints a summary of the options.
491 =item B<-i>[I<extension>]
494 specifies that files processed by the C<E<lt>E<gt>> construct are to be
495 edited in-place. It does this by renaming the input file, opening the
496 output file by the original name, and selecting that output file as the
497 default for print() statements. The extension, if supplied, is used to
498 modify the name of the old file to make a backup copy, following these
501 If no extension is supplied, no backup is made and the current file is
504 If the extension doesn't contain a C<*>, then it is appended to the
505 end of the current filename as a suffix. If the extension does
506 contain one or more C<*> characters, then each C<*> is replaced
507 with the current filename. In Perl terms, you could think of this
510 ($backup = $extension) =~ s/\*/$file_name/g;
512 This allows you to add a prefix to the backup file, instead of (or in
513 addition to) a suffix:
515 $ perl -pi'orig_*' -e 's/bar/baz/' fileA # backup to 'orig_fileA'
517 Or even to place backup copies of the original files into another
518 directory (provided the directory already exists):
520 $ perl -pi'old/*.orig' -e 's/bar/baz/' fileA # backup to 'old/fileA.orig'
522 These sets of one-liners are equivalent:
524 $ perl -pi -e 's/bar/baz/' fileA # overwrite current file
525 $ perl -pi'*' -e 's/bar/baz/' fileA # overwrite current file
527 $ perl -pi'.orig' -e 's/bar/baz/' fileA # backup to 'fileA.orig'
528 $ perl -pi'*.orig' -e 's/bar/baz/' fileA # backup to 'fileA.orig'
530 From the shell, saying
532 $ perl -p -i.orig -e "s/foo/bar/; ... "
534 is the same as using the program:
536 #!/usr/bin/perl -pi.orig
539 which is equivalent to
542 $extension = '.orig';
544 if ($ARGV ne $oldargv) {
545 if ($extension !~ /\*/) {
546 $backup = $ARGV . $extension;
549 ($backup = $extension) =~ s/\*/$ARGV/g;
551 rename($ARGV, $backup);
552 open(ARGVOUT, ">$ARGV");
559 print; # this prints to original filename
563 except that the B<-i> form doesn't need to compare $ARGV to $oldargv to
564 know when the filename has changed. It does, however, use ARGVOUT for
565 the selected filehandle. Note that STDOUT is restored as the default
566 output filehandle after the loop.
568 As shown above, Perl creates the backup file whether or not any output
569 is actually changed. So this is just a fancy way to copy files:
571 $ perl -p -i'/some/file/path/*' -e 1 file1 file2 file3...
573 $ perl -p -i'.orig' -e 1 file1 file2 file3...
575 You can use C<eof> without parentheses to locate the end of each input
576 file, in case you want to append to each file, or reset line numbering
577 (see example in L<perlfunc/eof>).
579 If, for a given file, Perl is unable to create the backup file as
580 specified in the extension then it will skip that file and continue on
581 with the next one (if it exists).
583 For a discussion of issues surrounding file permissions and B<-i>,
584 see L<perlfaq5/Why does Perl let me delete read-only files? Why does -i clobber protected files? Isn't this a bug in Perl?>.
586 You cannot use B<-i> to create directories or to strip extensions from
589 Perl does not expand C<~> in filenames, which is good, since some
590 folks use it for their backup files:
592 $ perl -pi~ -e 's/foo/bar/' file1 file2 file3...
594 Note that because B<-i> renames or deletes the original file before
595 creating a new file of the same name, Unix-style soft and hard links will
598 Finally, the B<-i> switch does not impede execution when no
599 files are given on the command line. In this case, no backup is made
600 (the original file cannot, of course, be determined) and processing
601 proceeds from STDIN to STDOUT as might be expected.
603 =item B<-I>I<directory>
606 Directories specified by B<-I> are prepended to the search path for
609 =item B<-l>[I<octnum>]
612 enables automatic line-ending processing. It has two separate
613 effects. First, it automatically chomps C<$/> (the input record
614 separator) when used with B<-n> or B<-p>. Second, it assigns C<$\>
615 (the output record separator) to have the value of I<octnum> so
616 that any print statements will have that separator added back on.
617 If I<octnum> is omitted, sets C<$\> to the current value of
618 C<$/>. For instance, to trim lines to 80 columns:
620 perl -lpe 'substr($_, 80) = ""'
622 Note that the assignment C<$\ = $/> is done when the switch is processed,
623 so the input record separator can be different than the output record
624 separator if the B<-l> switch is followed by a B<-0> switch:
626 gnufind / -print0 | perl -ln0e 'print "found $_" if -p'
628 This sets C<$\> to newline and then sets C<$/> to the null character.
630 =item B<-m>[B<->]I<module>
633 =item B<-M>[B<->]I<module>
635 =item B<-M>[B<->]I<'module ...'>
637 =item B<-[mM]>[B<->]I<module=arg[,arg]...>
639 B<-m>I<module> executes C<use> I<module> C<();> before executing your
642 B<-M>I<module> executes C<use> I<module> C<;> before executing your
643 program. You can use quotes to add extra code after the module name,
644 e.g., C<'-Mmodule qw(foo bar)'>.
646 If the first character after the B<-M> or B<-m> is a dash (C<->)
647 then the 'use' is replaced with 'no'.
649 A little builtin syntactic sugar means you can also say
650 B<-mmodule=foo,bar> or B<-Mmodule=foo,bar> as a shortcut for
651 C<'-Mmodule qw(foo bar)'>. This avoids the need to use quotes when
652 importing symbols. The actual code generated by B<-Mmodule=foo,bar> is
653 C<use module split(/,/,q{foo,bar})>. Note that the C<=> form
654 removes the distinction between B<-m> and B<-M>.
656 A consequence of this is that B<-MFoo=number> never does a version check
657 (unless C<Foo::import()> itself is set up to do a version check, which
658 could happen for example if Foo inherits from Exporter.)
663 causes Perl to assume the following loop around your program, which
664 makes it iterate over filename arguments somewhat like B<sed -n> or
669 ... # your program goes here
672 Note that the lines are not printed by default. See B<-p> to have
673 lines printed. If a file named by an argument cannot be opened for
674 some reason, Perl warns you about it and moves on to the next file.
676 Also note that C<< <> >> passes command line arguments to
677 L<perlfunc/open>, which doesn't necessarily interpret them as file names.
678 See L<perlop> for possible security implications.
680 Here is an efficient way to delete all files that haven't been modified for
683 find . -mtime +7 -print | perl -nle unlink
685 This is faster than using the B<-exec> switch of B<find> because you don't
686 have to start a process on every filename found. It does suffer from
687 the bug of mishandling newlines in pathnames, which you can fix if
688 you follow the example under B<-0>.
690 C<BEGIN> and C<END> blocks may be used to capture control before or after
691 the implicit program loop, just as in B<awk>.
696 causes Perl to assume the following loop around your program, which
697 makes it iterate over filename arguments somewhat like B<sed>:
702 ... # your program goes here
704 print or die "-p destination: $!\n";
707 If a file named by an argument cannot be opened for some reason, Perl
708 warns you about it, and moves on to the next file. Note that the
709 lines are printed automatically. An error occurring during printing is
710 treated as fatal. To suppress printing use the B<-n> switch. A B<-p>
711 overrides a B<-n> switch.
713 C<BEGIN> and C<END> blocks may be used to capture control before or after
714 the implicit loop, just as in B<awk>.
719 enables rudimentary switch parsing for switches on the command
720 line after the program name but before any filename arguments (or before
721 an argument of B<-->). Any switch found there is removed from @ARGV and sets the
722 corresponding variable in the Perl program. The following program
723 prints "1" if the program is invoked with a B<-xyz> switch, and "abc"
724 if it is invoked with B<-xyz=abc>.
727 if ($xyz) { print "$xyz\n" }
729 Do note that a switch like B<--help> creates the variable ${-help}, which is not compliant
730 with C<strict refs>. Also, when using this option on a script with
731 warnings enabled you may get a lot of spurious "used only once" warnings.
736 makes Perl use the PATH environment variable to search for the
737 program (unless the name of the program contains directory separators).
739 On some platforms, this also makes Perl append suffixes to the
740 filename while searching for it. For example, on Win32 platforms,
741 the ".bat" and ".cmd" suffixes are appended if a lookup for the
742 original name fails, and if the name does not already end in one
743 of those suffixes. If your Perl was compiled with DEBUGGING turned
744 on, using the -Dp switch to Perl shows how the search progresses.
746 Typically this is used to emulate #! startup on platforms that don't
747 support #!. Its also convenient when debugging a script that uses #!,
748 and is thus normally found by the shell's $PATH search mechanism.
750 This example works on many platforms that have a shell compatible with
754 eval 'exec /usr/bin/perl -wS $0 ${1+"$@"}'
755 if $running_under_some_shell;
757 The system ignores the first line and feeds the program to F</bin/sh>,
758 which proceeds to try to execute the Perl program as a shell script.
759 The shell executes the second line as a normal shell command, and thus
760 starts up the Perl interpreter. On some systems $0 doesn't always
761 contain the full pathname, so the B<-S> tells Perl to search for the
762 program if necessary. After Perl locates the program, it parses the
763 lines and ignores them because the variable $running_under_some_shell
764 is never true. If the program will be interpreted by csh, you will need
765 to replace C<${1+"$@"}> with C<$*>, even though that doesn't understand
766 embedded spaces (and such) in the argument list. To start up sh rather
767 than csh, some systems may have to replace the #! line with a line
768 containing just a colon, which will be politely ignored by Perl. Other
769 systems can't control that, and need a totally devious construct that
770 will work under any of B<csh>, B<sh>, or Perl, such as the following:
772 eval '(exit $?0)' && eval 'exec perl -wS $0 ${1+"$@"}'
773 & eval 'exec /usr/bin/perl -wS $0 $argv:q'
774 if $running_under_some_shell;
776 If the filename supplied contains directory separators (i.e., is an
777 absolute or relative pathname), and if that file is not found,
778 platforms that append file extensions will do so and try to look
779 for the file with those extensions added, one by one.
781 On DOS-like platforms, if the program does not contain directory
782 separators, it will first be searched for in the current directory
783 before being searched for on the PATH. On Unix platforms, the
784 program will be searched for strictly on the PATH.
789 Like B<-T>, but taint checks will issue warnings rather than fatal
790 errors. These warnings can be controlled normally with C<no warnings
793 B<NOTE: this is not a substitute for -T.> This is meant only to be
794 used as a temporary development aid while securing legacy code:
795 for real production code and for new secure code written from scratch
796 always use the real B<-T>.
801 forces "taint" checks to be turned on so you can test them. Ordinarily
802 these checks are done only when running setuid or setgid. It's a
803 good idea to turn them on explicitly for programs that run on behalf
804 of someone else whom you might not necessarily trust, such as CGI
805 programs or any internet servers you might write in Perl. See
806 L<perlsec> for details. For security reasons, this option must be
807 seen by Perl quite early; usually this means it must appear early
808 on the command line or in the #! line for systems which support
814 This obsolete switch causes Perl to dump core after compiling your
815 program. You can then in theory take this core dump and turn it
816 into an executable file by using the B<undump> program (not supplied).
817 This speeds startup at the expense of some disk space (which you
818 can minimize by stripping the executable). (Still, a "hello world"
819 executable comes out to about 200K on my machine.) If you want to
820 execute a portion of your program before dumping, use the dump()
821 operator instead. Note: availability of B<undump> is platform
822 specific and may not be available for a specific port of Perl.
827 allows Perl to do unsafe operations. Currently the only "unsafe"
828 operations are attempting to unlink directories while running as
829 superuser, and running setuid programs with fatal taint checks turned
830 into warnings. Note that the B<-w> switch (or the C<$^W> variable)
831 must be used along with this option to actually I<generate> the
832 taint-check warnings.
837 prints the version and patchlevel of your perl executable.
842 prints summary of the major perl configuration values and the current
845 =item B<-V:>I<configvar>
847 Prints to STDOUT the value of the named configuration variable(s),
848 with multiples when your configvar argument looks like a regex (has
849 non-letters). For example:
852 libc='/lib/libc-2.2.4.so';
854 libs='-lnsl -lgdbm -ldb -ldl -lm -lcrypt -lutil -lc';
855 libc='/lib/libc-2.2.4.so';
857 libpth='/usr/local/lib /lib /usr/lib';
858 libs='-lnsl -lgdbm -ldb -ldl -lm -lcrypt -lutil -lc';
860 libc='/lib/libc-2.2.4.so';
864 Additionally, extra colons can be used to control formatting. A
865 trailing colon suppresses the linefeed and terminator ';', allowing
866 you to embed queries into shell commands. (mnemonic: PATH separator
869 $ echo "compression-vars: " `perl -V:z.*: ` " are here !"
870 compression-vars: zcat='' zip='zip' are here !
872 A leading colon removes the 'name=' part of the response, this allows
873 you to map to the name you need. (mnemonic: empty label)
875 $ echo "goodvfork="`./perl -Ilib -V::usevfork`
878 Leading and trailing colons can be used together if you need
879 positional parameter values without the names. Note that in the case
880 below, the PERL_API params are returned in alphabetical order.
882 $ echo building_on `perl -V::osname: -V::PERL_API_.*:` now
883 building_on 'linux' '5' '1' '9' now
888 prints warnings about dubious constructs, such as variable names
889 that are mentioned only once and scalar variables that are used
890 before being set, redefined subroutines, references to undefined
891 filehandles or filehandles opened read-only that you are attempting
892 to write on, values used as a number that don't look like numbers,
893 using an array as though it were a scalar, if your subroutines
894 recurse more than 100 deep, and innumerable other things.
896 This switch really just enables the internal C<$^W> variable. You
897 can disable or promote into fatal errors specific warnings using
898 C<__WARN__> hooks, as described in L<perlvar> and L<perlfunc/warn>.
899 See also L<perldiag> and L<perltrap>. A new, fine-grained warning
900 facility is also available if you want to manipulate entire classes
901 of warnings; see L<warnings> or L<perllexwarn>.
906 Enables all warnings regardless of C<no warnings> or C<$^W>.
912 Disables all warnings regardless of C<use warnings> or C<$^W>.
918 =item B<-x>I<directory>
920 tells Perl that the program is embedded in a larger chunk of unrelated
921 ASCII text, such as in a mail message. Leading garbage will be
922 discarded until the first line that starts with #! and contains the
923 string "perl". Any meaningful switches on that line will be applied.
925 All references to line numbers by the program (warnings, errors, ...)
926 will treat the #! line as the first line.
927 Thus a warning on the 2nd line of the program (which is on the 100th
928 line in the file) will be reported as line 2, and not as line 100.
929 This can be overridden by using the #line directive.
930 (See L<perlsyn/"Plain-Old-Comments-(Not!)">)
932 If a directory name is specified, Perl will switch to that directory
933 before running the program. The B<-x> switch controls only the
934 disposal of leading garbage. The program must be terminated with
935 C<__END__> if there is trailing garbage to be ignored (the program
936 can process any or all of the trailing garbage via the DATA filehandle
939 The directory, if specified, must appear immediately following the B<-x>
940 with no intervening whitespace.
945 X<perl, environment variables>
952 Used if chdir has no argument.
957 Used if chdir has no argument and HOME is not set.
962 Used in executing subprocesses, and in finding the program if B<-S> is
968 A list of directories in which to look for Perl library
969 files before looking in the standard library and the current
970 directory. Any architecture-specific directories under the specified
971 locations are automatically included if they exist (this lookup
972 being done at interpreter startup time.)
974 If PERL5LIB is not defined, PERLLIB is used. Directories are separated
975 (like in PATH) by a colon on Unixish platforms and by a semicolon on
976 Windows (the proper path separator being given by the command C<perl
979 When running taint checks (either because the program was running setuid
980 or setgid, or the B<-T> or B<-t> switch was specified), neither variable
981 is used. The program should instead say:
983 use lib "/my/directory";
988 Command-line options (switches). Switches in this variable are taken
989 as if they were on every Perl command line. Only the B<-[CDIMUdmtwW]>
990 switches are allowed. When running taint checks (because the program
991 was running setuid or setgid, or the B<-T> switch was used), this
992 variable is ignored. If PERL5OPT begins with B<-T>, tainting will be
993 enabled, and any subsequent options ignored.
998 A space (or colon) separated list of PerlIO layers. If perl is built
999 to use PerlIO system for IO (the default) these layers effect perl's IO.
1001 It is conventional to start layer names with a colon e.g. C<:perlio> to
1002 emphasise their similarity to variable "attributes". But the code that parses
1003 layer specification strings (which is also used to decode the PERLIO
1004 environment variable) treats the colon as a separator.
1006 An unset or empty PERLIO is equivalent to the default set of layers for
1007 your platform, for example C<:unix:perlio> on Unix-like systems
1008 and C<:unix:crlf> on Windows and other DOS-like systems.
1010 The list becomes the default for I<all> perl's IO. Consequently only built-in
1011 layers can appear in this list, as external layers (such as :encoding()) need
1012 IO in order to load them!. See L<"open pragma"|open> for how to add external
1013 encodings as defaults.
1015 The layers that it makes sense to include in the PERLIO environment
1016 variable are briefly summarised below. For more details see L<PerlIO>.
1023 A pseudolayer that turns I<off> the C<:utf8> flag for the layer below.
1024 Unlikely to be useful on its own in the global PERLIO environment variable.
1025 You perhaps were thinking of C<:crlf:bytes> or C<:perlio:bytes>.
1030 A layer which does CRLF to "\n" translation distinguishing "text" and
1031 "binary" files in the manner of MS-DOS and similar operating systems.
1032 (It currently does I<not> mimic MS-DOS as far as treating of Control-Z
1033 as being an end-of-file marker.)
1038 A layer which implements "reading" of files by using C<mmap()> to
1039 make (whole) file appear in the process's address space, and then
1040 using that as PerlIO's "buffer".
1045 This is a re-implementation of "stdio-like" buffering written as a
1046 PerlIO "layer". As such it will call whatever layer is below it for
1047 its operations (typically C<:unix>).
1052 An experimental pseudolayer that removes the topmost layer.
1053 Use with the same care as is reserved for nitroglycerin.
1058 A pseudolayer that manipulates other layers. Applying the C<:raw>
1059 layer is equivalent to calling C<binmode($fh)>. It makes the stream
1060 pass each byte as-is without any translation. In particular CRLF
1061 translation, and/or :utf8 intuited from locale are disabled.
1063 Unlike in the earlier versions of Perl C<:raw> is I<not>
1064 just the inverse of C<:crlf> - other layers which would affect the
1065 binary nature of the stream are also removed or disabled.
1070 This layer provides PerlIO interface by wrapping system's ANSI C "stdio"
1071 library calls. The layer provides both buffering and IO.
1072 Note that C<:stdio> layer does I<not> do CRLF translation even if that
1073 is platforms normal behaviour. You will need a C<:crlf> layer above it
1079 Low level layer which calls C<read>, C<write> and C<lseek> etc.
1084 A pseudolayer that turns on a flag on the layer below to tell perl
1085 that output should be in utf8 and that input should be regarded as
1086 already in valid utf8 form. It does not check for validity and as such
1087 should be handled with caution for input. Generally C<:encoding(utf8)> is
1088 the best option when reading UTF-8 encoded data.
1093 On Win32 platforms this I<experimental> layer uses native "handle" IO
1094 rather than unix-like numeric file descriptor layer. Known to be
1095 buggy in this release.
1099 On all platforms the default set of layers should give acceptable results.
1101 For Unix platforms that will equivalent of "unix perlio" or "stdio".
1102 Configure is setup to prefer "stdio" implementation if system's library
1103 provides for fast access to the buffer, otherwise it uses the "unix perlio"
1106 On Win32 the default in this release is "unix crlf". Win32's "stdio"
1107 has a number of bugs/mis-features for perl IO which are somewhat
1108 C compiler vendor/version dependent. Using our own C<crlf> layer as
1109 the buffer avoids those issues and makes things more uniform.
1110 The C<crlf> layer provides CRLF to/from "\n" conversion as well as
1113 This release uses C<unix> as the bottom layer on Win32 and so still uses C
1114 compiler's numeric file descriptor routines. There is an experimental native
1115 C<win32> layer which is expected to be enhanced and should eventually be
1116 the default under Win32.
1118 The PERLIO environment variable is completely ignored when perl
1119 is run in taint mode.
1124 If set to the name of a file or device then certain operations of PerlIO
1125 sub-system will be logged to that file (opened as append). Typical uses
1128 PERLIO_DEBUG=/dev/tty perl script ...
1130 and Win32 approximate equivalent:
1132 set PERLIO_DEBUG=CON
1135 This functionality is disabled for setuid scripts and for scripts run
1141 A list of directories in which to look for Perl library
1142 files before looking in the standard library and the current directory.
1143 If PERL5LIB is defined, PERLLIB is not used.
1145 The PERLLIB environment variable is completely ignored when perl
1146 is run in taint mode.
1151 The command used to load the debugger code. The default is:
1153 BEGIN { require 'perl5db.pl' }
1155 The PERL5DB environment variable only used when perl is started with
1156 a bare B<-d> switch.
1158 =item PERL5DB_THREADED
1161 If set to a true value, indicates to the debugger that the code being
1162 debugged uses threads.
1164 =item PERL5SHELL (specific to the Win32 port)
1167 May be set to an alternative shell that perl must use internally for
1168 executing "backtick" commands or system(). Default is C<cmd.exe /x/d/c>
1169 on WindowsNT and C<command.com /c> on Windows95. The value is considered
1170 to be space-separated. Precede any character that needs to be protected
1171 (like a space or backslash) with a backslash.
1173 Note that Perl doesn't use COMSPEC for this purpose because
1174 COMSPEC has a high degree of variability among users, leading to
1175 portability concerns. Besides, perl can use a shell that may not be
1176 fit for interactive use, and setting COMSPEC to such a shell may
1177 interfere with the proper functioning of other programs (which usually
1178 look in COMSPEC to find a shell fit for interactive use).
1180 Before Perl 5.10.0 and 5.8.8, PERL5SHELL was not taint checked
1181 when running external commands. It is recommended that
1182 you explicitly set (or delete) C<$ENV{PERL5SHELL}> when running
1183 in taint mode under Windows.
1185 =item PERL_ALLOW_NON_IFS_LSP (specific to the Win32 port)
1186 X<PERL_ALLOW_NON_IFS_LSP>
1188 Set to 1 to allow the use of non-IFS compatible LSP's.
1189 Perl normally searches for an IFS-compatible LSP because this is required
1190 for its emulation of Windows sockets as real filehandles. However, this may
1191 cause problems if you have a firewall such as McAfee Guardian which requires
1192 all applications to use its LSP which is not IFS-compatible, because clearly
1193 Perl will normally avoid using such an LSP.
1194 Setting this environment variable to 1 means that Perl will simply use the
1195 first suitable LSP enumerated in the catalog, which keeps McAfee Guardian
1196 happy (and in that particular case Perl still works too because McAfee
1197 Guardian's LSP actually plays some other games which allow applications
1198 requiring IFS compatibility to work).
1200 =item PERL_DEBUG_MSTATS
1201 X<PERL_DEBUG_MSTATS>
1203 Relevant only if perl is compiled with the malloc included with the perl
1204 distribution (that is, if C<perl -V:d_mymalloc> is 'define').
1205 If set, this causes memory statistics to be dumped after execution. If set
1206 to an integer greater than one, also causes memory statistics to be dumped
1209 =item PERL_DESTRUCT_LEVEL
1210 X<PERL_DESTRUCT_LEVEL>
1212 Relevant only if your perl executable was built with B<-DDEBUGGING>,
1213 this controls the behavior of global destruction of objects and other
1214 references. See L<perlhack/PERL_DESTRUCT_LEVEL> for more information.
1216 =item PERL_DL_NONLAZY
1219 Set to one to have perl resolve B<all> undefined symbols when it loads
1220 a dynamic library. The default behaviour is to resolve symbols when
1221 they are used. Setting this variable is useful during testing of
1222 extensions as it ensures that you get an error on misspelled function
1223 names even if the test suite doesn't call it.
1228 If using the C<encoding> pragma without an explicit encoding name, the
1229 PERL_ENCODING environment variable is consulted for an encoding name.
1231 =item PERL_HASH_SEED
1234 (Since Perl 5.8.1.) Used to randomise perl's internal hash function.
1235 To emulate the pre-5.8.1 behaviour, set to an integer (zero means
1236 exactly the same order as 5.8.0). "Pre-5.8.1" means, among other
1237 things, that hash keys will always have the same ordering between
1238 different runs of perl.
1240 Most hashes return elements in the same order as Perl 5.8.0 by default.
1241 On a hash by hash basis, if pathological data is detected during a hash
1242 key insertion, then that hash will switch to an alternative random hash
1245 The default behaviour is to randomise unless the PERL_HASH_SEED is set.
1246 If perl has been compiled with C<-DUSE_HASH_SEED_EXPLICIT>, the default
1247 behaviour is B<not> to randomise unless the PERL_HASH_SEED is set.
1249 If PERL_HASH_SEED is unset or set to a non-numeric string, perl uses
1250 the pseudorandom seed supplied by the operating system and libraries.
1252 B<Please note that the hash seed is sensitive information>. Hashes are
1253 randomized to protect against local and remote attacks against Perl
1254 code. By manually setting a seed this protection may be partially or
1257 See L<perlsec/"Algorithmic Complexity Attacks"> and
1258 L</PERL_HASH_SEED_DEBUG> for more information.
1260 =item PERL_HASH_SEED_DEBUG
1261 X<PERL_HASH_SEED_DEBUG>
1263 (Since Perl 5.8.1.) Set to one to display (to STDERR) the value of
1264 the hash seed at the beginning of execution. This, combined with
1265 L</PERL_HASH_SEED> is intended to aid in debugging nondeterministic
1266 behavior caused by hash randomization.
1268 B<Note that the hash seed is sensitive information>: by knowing it one
1269 can craft a denial-of-service attack against Perl code, even remotely,
1270 see L<perlsec/"Algorithmic Complexity Attacks"> for more information.
1271 B<Do not disclose the hash seed> to people who don't need to know it.
1272 See also hash_seed() of L<Hash::Util>.
1277 If your perl was configured with C<-Accflags=-DPERL_MEM_LOG>, setting
1278 the environment variable C<PERL_MEM_LOG> enables logging debug
1279 messages. The value has the form C<< <number>[m][s][t] >>, where
1280 C<number> is the filedescriptor number you want to write to (2 is
1281 default), and the combination of letters specifies that you want
1282 information about (m)emory and/or (s)v, optionally with
1283 (t)imestamps. For example C<PERL_MEM_LOG=1mst> will log all
1284 information to stdout. You can write to other opened filedescriptors
1285 too, in a variety of ways;
1287 bash$ 3>foo3 PERL_MEM_LOG=3m perl ...
1289 =item PERL_ROOT (specific to the VMS port)
1292 A translation concealed rooted logical name that contains perl and the
1293 logical device for the @INC path on VMS only. Other logical names that
1294 affect perl on VMS include PERLSHR, PERL_ENV_TABLES, and
1295 SYS$TIMEZONE_DIFFERENTIAL but are optional and discussed further in
1296 L<perlvms> and in F<README.vms> in the Perl source distribution.
1301 In Perls 5.8.1 and later. If set to C<unsafe> the pre-Perl-5.8.0
1302 signals behaviour (immediate but unsafe) is restored. If set to
1303 C<safe> the safe (or deferred) signals are used.
1304 See L<perlipc/"Deferred Signals (Safe Signals)">.
1309 Equivalent to the B<-C> command-line switch. Note that this is not
1310 a boolean variable. Setting this to C<"1"> is not the right way to
1311 "enable Unicode" (whatever that would mean). You can use C<"0"> to
1312 "disable Unicode", though (or alternatively unset PERL_UNICODE in
1313 your shell before starting Perl). See the description of the C<-C>
1314 switch for more information.
1316 =item SYS$LOGIN (specific to the VMS port)
1319 Used if chdir has no argument and HOME and LOGDIR are not set.
1323 Perl also has environment variables that control how Perl handles data
1324 specific to particular natural languages. See L<perllocale>.
1326 Apart from these, Perl uses no other environment variables, except
1327 to make them available to the program being executed, and to child
1328 processes. However, programs running setuid would do well to execute
1329 the following lines before doing anything else, just to keep people
1332 $ENV{PATH} = '/bin:/usr/bin'; # or whatever you need
1333 $ENV{SHELL} = '/bin/sh' if exists $ENV{SHELL};
1334 delete @ENV{qw(IFS CDPATH ENV BASH_ENV)};