summaryrefslogtreecommitdiff
path: root/pod/perlrun.pod
blob: 1e83bdc114c1bbddbba628ec817ce8cb5d222dff (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
=head1 NAME

perlrun - how to execute the Perl interpreter

=head1 SYNOPSIS

B<perl>	S<[ B<-sTtuUWX> ]>
	S<[ B<-hv> ] [ B<-V>[:I<configvar>] ]>
	S<[ B<-cw> ] [ B<-d>[B<t>][:I<debugger>] ] [ B<-D>[I<number/list>] ]>
	S<[ B<-pna> ] [ B<-F>I<pattern> ] [ B<-l>[I<octal>] ] [ B<-0>[I<octal/hexadecimal>] ]>
	S<[ B<-I>I<dir> ] [ B<-m>[B<->]I<module> ] [ B<-M>[B<->]I<'module...'> ]>
	S<[ B<-P> ]>
	S<[ B<-S> ]>
	S<[ B<-x>[I<dir>] ]>
	S<[ B<-i>[I<extension>] ]>
	S<[ B<-e> I<'command'> ] [ B<--> ] [ I<programfile> ] [ I<argument> ]...>
	S<[ B<-A [I<assertions>] >]>
	S<[ B<-C [I<number/list>] >]>

=head1 DESCRIPTION

The normal way to run a Perl program is by making it directly
executable, or else by passing the name of the source file as an
argument on the command line.  (An interactive Perl environment
is also possible--see L<perldebug> for details on how to do that.)
Upon startup, Perl looks for your program in one of the following
places:

=over 4

=item 1.

Specified line by line via B<-e> switches on the command line.

=item 2.

Contained in the file specified by the first filename on the command line.
(Note that systems supporting the #! notation invoke interpreters this
way. See L<Location of Perl>.)

=item 3.

Passed in implicitly via standard input.  This works only if there are
no filename arguments--to pass arguments to a STDIN-read program you
must explicitly specify a "-" for the program name.

=back

With methods 2 and 3, Perl starts parsing the input file from the
beginning, unless you've specified a B<-x> switch, in which case it
scans for the first line starting with #! and containing the word
"perl", and starts there instead.  This is useful for running a program
embedded in a larger message.  (In this case you would indicate the end
of the program using the C<__END__> token.)

The #! line is always examined for switches as the line is being
parsed.  Thus, if you're on a machine that allows only one argument
with the #! line, or worse, doesn't even recognize the #! line, you
still can get consistent switch behavior regardless of how Perl was
invoked, even if B<-x> was used to find the beginning of the program.

Because historically some operating systems silently chopped off
kernel interpretation of the #! line after 32 characters, some
switches may be passed in on the command line, and some may not;
you could even get a "-" without its letter, if you're not careful.
You probably want to make sure that all your switches fall either
before or after that 32-character boundary.  Most switches don't
actually care if they're processed redundantly, but getting a "-"
instead of a complete switch could cause Perl to try to execute
standard input instead of your program.  And a partial B<-I> switch
could also cause odd results.

Some switches do care if they are processed twice, for instance
combinations of B<-l> and B<-0>.  Either put all the switches after
the 32-character boundary (if applicable), or replace the use of
B<-0>I<digits> by C<BEGIN{ $/ = "\0digits"; }>.

Parsing of the #! switches starts wherever "perl" is mentioned in the line.
The sequences "-*" and "- " are specifically ignored so that you could,
if you were so inclined, say

    #!/bin/sh -- # -*- perl -*- -p
    eval 'exec perl -wS $0 ${1+"$@"}'
        if $running_under_some_shell;

to let Perl see the B<-p> switch.

A similar trick involves the B<env> program, if you have it.

    #!/usr/bin/env perl

The examples above use a relative path to the perl interpreter,
getting whatever version is first in the user's path.  If you want
a specific version of Perl, say, perl5.005_57, you should place
that directly in the #! line's path.

If the #! line does not contain the word "perl", the program named after
the #! is executed instead of the Perl interpreter.  This is slightly
bizarre, but it helps people on machines that don't do #!, because they
can tell a program that their SHELL is F</usr/bin/perl>, and Perl will then
dispatch the program to the correct interpreter for them.

After locating your program, Perl compiles the entire program to an
internal form.  If there are any compilation errors, execution of the
program is not attempted.  (This is unlike the typical shell script,
which might run part-way through before finding a syntax error.)

If the program is syntactically correct, it is executed.  If the program
runs off the end without hitting an exit() or die() operator, an implicit
C<exit(0)> is provided to indicate successful completion.

=head2 #! and quoting on non-Unix systems

Unix's #! technique can be simulated on other systems:

=over 4

=item OS/2

Put

    extproc perl -S -your_switches

as the first line in C<*.cmd> file (B<-S> due to a bug in cmd.exe's
`extproc' handling).

=item MS-DOS

Create a batch file to run your program, and codify it in
C<ALTERNATE_SHEBANG> (see the F<dosish.h> file in the source
distribution for more information).

=item Win95/NT

The Win95/NT installation, when using the ActiveState installer for Perl,
will modify the Registry to associate the F<.pl> extension with the perl
interpreter.  If you install Perl by other means (including building from
the sources), you may have to modify the Registry yourself.  Note that
this means you can no longer tell the difference between an executable
Perl program and a Perl library file.

=item Macintosh

Under "Classic" MacOS, a perl program will have the appropriate Creator and
Type, so that double-clicking them will invoke the MacPerl application.
Under Mac OS X, clickable apps can be made from any C<#!> script using Wil
Sanchez' DropScript utility: http://www.wsanchez.net/software/ .

=item VMS

Put

    $ perl -mysw 'f$env("procedure")' 'p1' 'p2' 'p3' 'p4' 'p5' 'p6' 'p7' 'p8' !
    $ exit++ + ++$status != 0 and $exit = $status = undef;

at the top of your program, where B<-mysw> are any command line switches you
want to pass to Perl.  You can now invoke the program directly, by saying
C<perl program>, or as a DCL procedure, by saying C<@program> (or implicitly
via F<DCL$PATH> by just using the name of the program).

This incantation is a bit much to remember, but Perl will display it for
you if you say C<perl "-V:startperl">.

=back

Command-interpreters on non-Unix systems have rather different ideas
on quoting than Unix shells.  You'll need to learn the special
characters in your command-interpreter (C<*>, C<\> and C<"> are
common) and how to protect whitespace and these characters to run
one-liners (see B<-e> below).

On some systems, you may have to change single-quotes to double ones,
which you must I<not> do on Unix or Plan 9 systems.  You might also
have to change a single % to a %%.

For example:

    # Unix
    perl -e 'print "Hello world\n"'

    # MS-DOS, etc.
    perl -e "print \"Hello world\n\""

    # Macintosh
    print "Hello world\n"
     (then Run "Myscript" or Shift-Command-R)

    # VMS
    perl -e "print ""Hello world\n"""

The problem is that none of this is reliable: it depends on the
command and it is entirely possible neither works.  If B<4DOS> were
the command shell, this would probably work better:

    perl -e "print <Ctrl-x>"Hello world\n<Ctrl-x>""

B<CMD.EXE> in Windows NT slipped a lot of standard Unix functionality in
when nobody was looking, but just try to find documentation for its
quoting rules.

Under the Macintosh, it depends which environment you are using.  The MacPerl
shell, or MPW, is much like Unix shells in its support for several
quoting variants, except that it makes free use of the Macintosh's non-ASCII
characters as control characters.

There is no general solution to all of this.  It's just a mess.

=head2 Location of Perl

It may seem obvious to say, but Perl is useful only when users can
easily find it.  When possible, it's good for both F</usr/bin/perl>
and F</usr/local/bin/perl> to be symlinks to the actual binary.  If
that can't be done, system administrators are strongly encouraged
to put (symlinks to) perl and its accompanying utilities into a
directory typically found along a user's PATH, or in some other
obvious and convenient place.

In this documentation, C<#!/usr/bin/perl> on the first line of the program
will stand in for whatever method works on your system.  You are
advised to use a specific path if you care about a specific version.

    #!/usr/local/bin/perl5.00554

or if you just want to be running at least version, place a statement
like this at the top of your program:

    use 5.005_54;

=head2 Command Switches

As with all standard commands, a single-character switch may be
clustered with the following switch, if any.

    #!/usr/bin/perl -spi.orig	# same as -s -p -i.orig

Switches include:

=over 5

=item B<-0>[I<octal/hexadecimal>]

specifies the input record separator (C<$/>) as an octal or
hexadecimal number.  If there are no digits, the null character is the
separator.  Other switches may precede or follow the digits.  For
example, if you have a version of B<find> which can print filenames
terminated by the null character, you can say this:

    find . -name '*.orig' -print0 | perl -n0e unlink

The special value 00 will cause Perl to slurp files in paragraph mode.
The value 0777 will cause Perl to slurp files whole because there is no
legal byte with that value.

If you want to specify any Unicode character, use the hexadecimal
format: C<-0xHHH...>, where the C<H> are valid hexadecimal digits.
(This means that you cannot use the C<-x> with a directory name that
consists of hexadecimal digits.)

=item B<-A [I<assertions>]>

Activates the assertions given after the switch as a comma-separated
list of assertion names. If no assertion name is given, activates all
assertions. See L<assertions>.

=item B<-a>

turns on autosplit mode when used with a B<-n> or B<-p>.  An implicit
split command to the @F array is done as the first thing inside the
implicit while loop produced by the B<-n> or B<-p>.

    perl -ane 'print pop(@F), "\n";'

is equivalent to

    while (<>) {
	@F = split(' ');
	print pop(@F), "\n";
    }

An alternate delimiter may be specified using B<-F>.

=item B<-C [I<number/list>]>

The C<-C> flag controls some Unicode of the Perl Unicode features.

As of 5.8.1, the C<-C> can be followed either by a number or a list
of option letters.  The letters, their numeric values, and effects
are as follows; listing the letters is equal to summing the numbers.

    I     1   STDIN is assumed to be in UTF-8
    O     2   STDOUT will be in UTF-8
    E     4   STDERR will be in UTF-8
    S     7   I + O + E
    i     8   UTF-8 is the default PerlIO layer for input streams
    o    16   UTF-8 is the default PerlIO layer for output streams
    D    24   i + o
    A    32   the @ARGV elements are expected to be strings encoded
              in UTF-8
    L    64   normally the "IOEioA" are unconditional,
              the L makes them conditional on the locale environment
              variables (the LC_ALL, LC_TYPE, and LANG, in the order
              of decreasing precedence) -- if the variables indicate
              UTF-8, then the selected "IOEioA" are in effect

For example, C<-COE> and C<-C6> will both turn on UTF-8-ness on both
STDOUT and STDERR.  Repeating letters is just redundant, not cumulative
nor toggling.

The C<io> options mean that any subsequent open() (or similar I/O
operations) will have the C<:utf8> PerlIO layer implicitly applied
to them, in other words, UTF-8 is expected from any input stream,
and UTF-8 is produced to any output stream.  This is just the default,
with explicit layers in open() and with binmode() one can manipulate
streams as usual.

C<-C> on its own (not followed by any number or option list), or the
empty string C<""> for the C<PERL_UNICODE> environment variable, has the
same effect as C<-CSDL>.  In other words, the standard I/O handles and
the default C<open()> layer are UTF-8-fied B<but> only if the locale
environment variables indicate a UTF-8 locale.  This behaviour follows
the I<implicit> (and problematic) UTF-8 behaviour of Perl 5.8.0.

You can use C<-C0> (or C<"0"> for C<PERL_UNICODE>) to explicitly
disable all the above Unicode features.

The read-only magic variable C<${^UNICODE}> reflects the numeric value
of this setting.  This is variable is set during Perl startup and is
thereafter read-only.  If you want runtime effects, use the three-arg
open() (see L<perlfunc/open>), the two-arg binmode() (see L<perlfunc/binmode>),
and the C<open> pragma (see L<open>).

(In Perls earlier than 5.8.1 the C<-C> switch was a Win32-only switch
that enabled the use of Unicode-aware "wide system call" Win32 APIs.
This feature was practically unused, however, and the command line
switch was therefore "recycled".)

=item B<-c>

causes Perl to check the syntax of the program and then exit without
executing it.  Actually, it I<will> execute C<BEGIN>, C<CHECK>, and
C<use> blocks, because these are considered as occurring outside the
execution of your program.  C<INIT> and C<END> blocks, however, will
be skipped.

=item B<-d>

=item B<-dt>

runs the program under the Perl debugger.  See L<perldebug>.
If B<t> is specified, it indicates to the debugger that threads
will be used in the code being debugged.

=item B<-d:>I<foo[=bar,baz]>

=item B<-dt:>I<foo[=bar,baz]>

runs the program under the control of a debugging, profiling, or
tracing module installed as Devel::foo. E.g., B<-d:DProf> executes
the program using the Devel::DProf profiler.  As with the B<-M>
flag, options may be passed to the Devel::foo package where they
will be received and interpreted by the Devel::foo::import routine.
The comma-separated list of options must follow a C<=> character.
If B<t> is specified, it indicates to the debugger that threads
will be used in the code being debugged.
See L<perldebug>.

=item B<-D>I<letters>

=item B<-D>I<number>

sets debugging flags.  To watch how it executes your program, use
B<-Dtls>.  (This works only if debugging is compiled into your
Perl.)  Another nice value is B<-Dx>, which lists your compiled
syntax tree.  And B<-Dr> displays compiled regular expressions;
the format of the output is explained in L<perldebguts>.

As an alternative, specify a number instead of list of letters (e.g.,
B<-D14> is equivalent to B<-Dtls>):

        1  p  Tokenizing and parsing (with v, displays parse stack)
        2  s  Stack snapshots (with v, displays all stacks)
        4  l  Context (loop) stack processing
        8  t  Trace execution
       16  o  Method and overloading resolution
       32  c  String/numeric conversions
       64  P  Print profiling info, preprocessor command for -P, source file input state
      128  m  Memory allocation
      256  f  Format processing
      512  r  Regular expression parsing and execution
     1024  x  Syntax tree dump
     2048  u  Tainting checks
     4096     (Obsolete, previously used for LEAKTEST)
     8192  H  Hash dump -- usurps values()
    16384  X  Scratchpad allocation
    32768  D  Cleaning up
    65536  S  Thread synchronization
   131072  T  Tokenising
   262144  R  Include reference counts of dumped variables (eg when using -Ds)
   524288  J  Do not s,t,P-debug (Jump over) opcodes within package DB
  1048576  v  Verbose: use in conjunction with other flags
  2097152  C  Copy On Write
  4194304  A  Consistency checks on internal structures
  8388608  q  quiet - currently only suppresses the "EXECUTING" message

All these flags require B<-DDEBUGGING> when you compile the Perl
executable (but see L<Devel::Peek>, L<re> which may change this).
See the F<INSTALL> file in the Perl source distribution
for how to do this.  This flag is automatically set if you include B<-g>
option when C<Configure> asks you about optimizer/debugger flags.

If you're just trying to get a print out of each line of Perl code
as it executes, the way that C<sh -x> provides for shell scripts,
you can't use Perl's B<-D> switch.  Instead do this

  # If you have "env" utility
  env=PERLDB_OPTS="NonStop=1 AutoTrace=1 frame=2" perl -dS program

  # Bourne shell syntax
  $ PERLDB_OPTS="NonStop=1 AutoTrace=1 frame=2" perl -dS program

  # csh syntax
  % (setenv PERLDB_OPTS "NonStop=1 AutoTrace=1 frame=2"; perl -dS program)

See L<perldebug> for details and variations.

=item B<-e> I<commandline>

may be used to enter one line of program.  If B<-e> is given, Perl
will not look for a filename in the argument list.  Multiple B<-e>
commands may be given to build up a multi-line script.  Make sure
to use semicolons where you would in a normal program.

=item B<-F>I<pattern>

specifies the pattern to split on if B<-a> is also in effect.  The
pattern may be surrounded by C<//>, C<"">, or C<''>, otherwise it will be
put in single quotes.

=item B<-h>

prints a summary of the options.

=item B<-i>[I<extension>]

specifies that files processed by the C<E<lt>E<gt>> construct are to be
edited in-place.  It does this by renaming the input file, opening the
output file by the original name, and selecting that output file as the
default for print() statements.  The extension, if supplied, is used to
modify the name of the old file to make a backup copy, following these
rules:

If no extension is supplied, no backup is made and the current file is
overwritten.

If the extension doesn't contain a C<*>, then it is appended to the
end of the current filename as a suffix.  If the extension does
contain one or more C<*> characters, then each C<*> is replaced
with the current filename.  In Perl terms, you could think of this
as:

    ($backup = $extension) =~ s/\*/$file_name/g;

This allows you to add a prefix to the backup file, instead of (or in
addition to) a suffix:

    $ perl -pi'orig_*' -e 's/bar/baz/' fileA	# backup to 'orig_fileA'

Or even to place backup copies of the original files into another
directory (provided the directory already exists):

    $ perl -pi'old/*.orig' -e 's/bar/baz/' fileA # backup to 'old/fileA.orig'

These sets of one-liners are equivalent:

    $ perl -pi -e 's/bar/baz/' fileA		# overwrite current file
    $ perl -pi'*' -e 's/bar/baz/' fileA		# overwrite current file

    $ perl -pi'.orig' -e 's/bar/baz/' fileA	# backup to 'fileA.orig'
    $ perl -pi'*.orig' -e 's/bar/baz/' fileA	# backup to 'fileA.orig'

From the shell, saying

    $ perl -p -i.orig -e "s/foo/bar/; ... "

is the same as using the program:

    #!/usr/bin/perl -pi.orig
    s/foo/bar/;

which is equivalent to

    #!/usr/bin/perl
    $extension = '.orig';
    LINE: while (<>) {
	if ($ARGV ne $oldargv) {
	    if ($extension !~ /\*/) {
		$backup = $ARGV . $extension;
	    }
	    else {
		($backup = $extension) =~ s/\*/$ARGV/g;
	    }
	    rename($ARGV, $backup);
	    open(ARGVOUT, ">$ARGV");
	    select(ARGVOUT);
	    $oldargv = $ARGV;
	}
	s/foo/bar/;
    }
    continue {
	print;	# this prints to original filename
    }
    select(STDOUT);

except that the B<-i> form doesn't need to compare $ARGV to $oldargv to
know when the filename has changed.  It does, however, use ARGVOUT for
the selected filehandle.  Note that STDOUT is restored as the default
output filehandle after the loop.

As shown above, Perl creates the backup file whether or not any output
is actually changed.  So this is just a fancy way to copy files:

    $ perl -p -i'/some/file/path/*' -e 1 file1 file2 file3...
or
    $ perl -p -i'.orig' -e 1 file1 file2 file3...

You can use C<eof> without parentheses to locate the end of each input
file, in case you want to append to each file, or reset line numbering
(see example in L<perlfunc/eof>).

If, for a given file, Perl is unable to create the backup file as
specified in the extension then it will skip that file and continue on
with the next one (if it exists).

For a discussion of issues surrounding file permissions and B<-i>,
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?>.

You cannot use B<-i> to create directories or to strip extensions from
files.

Perl does not expand C<~> in filenames, which is good, since some
folks use it for their backup files:

    $ perl -pi~ -e 's/foo/bar/' file1 file2 file3...

Note that because B<-i> renames or deletes the original file before
creating a new file of the same name, UNIX-style soft and hard links will
not be preserved.

Finally, the B<-i> switch does not impede execution when no
files are given on the command line.  In this case, no backup is made
(the original file cannot, of course, be determined) and processing
proceeds from STDIN to STDOUT as might be expected.

=item B<-I>I<directory>

Directories specified by B<-I> are prepended to the search path for
modules (C<@INC>), and also tells the C preprocessor where to search for
include files.  The C preprocessor is invoked with B<-P>; by default it
searches /usr/include and /usr/lib/perl.

=item B<-l>[I<octnum>]

enables automatic line-ending processing.  It has two separate
effects.  First, it automatically chomps C<$/> (the input record
separator) when used with B<-n> or B<-p>.  Second, it assigns C<$\>
(the output record separator) to have the value of I<octnum> so
that any print statements will have that separator added back on.
If I<octnum> is omitted, sets C<$\> to the current value of
C<$/>.  For instance, to trim lines to 80 columns:

    perl -lpe 'substr($_, 80) = ""'

Note that the assignment C<$\ = $/> is done when the switch is processed,
so the input record separator can be different than the output record
separator if the B<-l> switch is followed by a B<-0> switch:

    gnufind / -print0 | perl -ln0e 'print "found $_" if -p'

This sets C<$\> to newline and then sets C<$/> to the null character.

=item B<-m>[B<->]I<module>

=item B<-M>[B<->]I<module>

=item B<-M>[B<->]I<'module ...'>

=item B<-[mM]>[B<->]I<module=arg[,arg]...>

B<-m>I<module> executes C<use> I<module> C<();> before executing your
program.

B<-M>I<module> executes C<use> I<module> C<;> before executing your
program.  You can use quotes to add extra code after the module name,
e.g., C<'-Mmodule qw(foo bar)'>.

If the first character after the B<-M> or B<-m> is a dash (C<->)
then the 'use' is replaced with 'no'.

A little builtin syntactic sugar means you can also say
B<-mmodule=foo,bar> or B<-Mmodule=foo,bar> as a shortcut for
C<'-Mmodule qw(foo bar)'>.  This avoids the need to use quotes when
importing symbols.  The actual code generated by B<-Mmodule=foo,bar> is
C<use module split(/,/,q{foo,bar})>.  Note that the C<=> form
removes the distinction between B<-m> and B<-M>.

=item B<-n>

causes Perl to assume the following loop around your program, which
makes it iterate over filename arguments somewhat like B<sed -n> or
B<awk>:

  LINE:
    while (<>) {
	...		# your program goes here
    }

Note that the lines are not printed by default.  See B<-p> to have
lines printed.  If a file named by an argument cannot be opened for
some reason, Perl warns you about it and moves on to the next file.

Here is an efficient way to delete all files that haven't been modified for
at least a week:

    find . -mtime +7 -print | perl -nle unlink

This is faster than using the B<-exec> switch of B<find> because you don't
have to start a process on every filename found.  It does suffer from
the bug of mishandling newlines in pathnames, which you can fix if
you follow the example under B<-0>.

C<BEGIN> and C<END> blocks may be used to capture control before or after
the implicit program loop, just as in B<awk>.

=item B<-p>

causes Perl to assume the following loop around your program, which
makes it iterate over filename arguments somewhat like B<sed>:


  LINE:
    while (<>) {
	...		# your program goes here
    } continue {
	print or die "-p destination: $!\n";
    }

If a file named by an argument cannot be opened for some reason, Perl
warns you about it, and moves on to the next file.  Note that the
lines are printed automatically.  An error occurring during printing is
treated as fatal.  To suppress printing use the B<-n> switch.  A B<-p>
overrides a B<-n> switch.

C<BEGIN> and C<END> blocks may be used to capture control before or after
the implicit loop, just as in B<awk>.

=item B<-P>

B<NOTE: Use of -P is strongly discouraged because of its inherent
problems, including poor portability.>

This option causes your program to be run through the C preprocessor before
compilation by Perl.  Because both comments and B<cpp> directives begin
with the # character, you should avoid starting comments with any words
recognized by the C preprocessor such as C<"if">, C<"else">, or C<"define">.

If you're considering using C<-P>, you might also want to look at the
Filter::cpp module from CPAN.

The problems of -P include, but are not limited to:

=over 10

=item *

The C<#!> line is stripped, so any switches there don't apply.

=item *

A C<-P> on a C<#!> line doesn't work.

=item *

B<All> lines that begin with (whitespace and) a C<#> but
do not look like cpp commands, are stripped, including anything
inside Perl strings, regular expressions, and here-docs .

=item *

In some platforms the C preprocessor knows too much: it knows about
the C++ -style until-end-of-line comments starting with C<"//">.
This will cause problems with common Perl constructs like

    s/foo//;

because after -P this will became illegal code

    s/foo

The workaround is to use some other quoting separator than C<"/">,
like for example C<"!">:

    s!foo!!;



=item *

It requires not only a working C preprocessor but also a working
F<sed>.  If not on UNIX, you are probably out of luck on this.

=item *

Script line numbers are not preserved.

=item *

The C<-x> does not work with C<-P>.

=back

=item B<-s>

enables rudimentary switch parsing for switches on the command
line after the program name but before any filename arguments (or before
an argument of B<-->).  This means you can have switches with two leading
dashes (B<--help>).  Any switch found there is removed from @ARGV and sets the
corresponding variable in the Perl program.  The following program
prints "1" if the program is invoked with a B<-xyz> switch, and "abc"
if it is invoked with B<-xyz=abc>.

    #!/usr/bin/perl -s
    if ($xyz) { print "$xyz\n" }

Do note that B<--help> creates the variable ${-help}, which is not compliant
with C<strict refs>.  Also, when using this option on a script with
warnings enabled you may get a lot of spurious "used only once" warnings.

=item B<-S>

makes Perl use the PATH environment variable to search for the
program (unless the name of the program contains directory separators).

On some platforms, this also makes Perl append suffixes to the
filename while searching for it.  For example, on Win32 platforms,
the ".bat" and ".cmd" suffixes are appended if a lookup for the
original name fails, and if the name does not already end in one
of those suffixes.  If your Perl was compiled with DEBUGGING turned
on, using the -Dp switch to Perl shows how the search progresses.

Typically this is used to emulate #! startup on platforms that don't
support #!.  Its also convenient when debugging a script that uses #!,
and is thus normally found by the shell's $PATH search mechanism.

This example works on many platforms that have a shell compatible with
Bourne shell:

    #!/usr/bin/perl
    eval 'exec /usr/bin/perl -wS $0 ${1+"$@"}'
	    if $running_under_some_shell;

The system ignores the first line and feeds the program to F</bin/sh>,
which proceeds to try to execute the Perl program as a shell script.
The shell executes the second line as a normal shell command, and thus
starts up the Perl interpreter.  On some systems $0 doesn't always
contain the full pathname, so the B<-S> tells Perl to search for the
program if necessary.  After Perl locates the program, it parses the
lines and ignores them because the variable $running_under_some_shell
is never true.  If the program will be interpreted by csh, you will need
to replace C<${1+"$@"}> with C<$*>, even though that doesn't understand
embedded spaces (and such) in the argument list.  To start up sh rather
than csh, some systems may have to replace the #! line with a line
containing just a colon, which will be politely ignored by Perl.  Other
systems can't control that, and need a totally devious construct that
will work under any of B<csh>, B<sh>, or Perl, such as the following:

	eval '(exit $?0)' && eval 'exec perl -wS $0 ${1+"$@"}'
	& eval 'exec /usr/bin/perl -wS $0 $argv:q'
		if $running_under_some_shell;

If the filename supplied contains directory separators (i.e., is an
absolute or relative pathname), and if that file is not found,
platforms that append file extensions will do so and try to look
for the file with those extensions added, one by one.

On DOS-like platforms, if the program does not contain directory
separators, it will first be searched for in the current directory
before being searched for on the PATH.  On Unix platforms, the
program will be searched for strictly on the PATH.

=item B<-t>

Like B<-T>, but taint checks will issue warnings rather than fatal
errors.  These warnings can be controlled normally with C<no warnings
qw(taint)>.

B<NOTE: this is not a substitute for -T.> This is meant only to be
used as a temporary development aid while securing legacy code:
for real production code and for new secure code written from scratch
always use the real B<-T>.

=item B<-T>

forces "taint" checks to be turned on so you can test them.  Ordinarily
these checks are done only when running setuid or setgid.  It's a
good idea to turn them on explicitly for programs that run on behalf
of someone else whom you might not necessarily trust, such as CGI
programs or any internet servers you might write in Perl.  See
L<perlsec> for details.  For security reasons, this option must be
seen by Perl quite early; usually this means it must appear early
on the command line or in the #! line for systems which support
that construct.

=item B<-u>

This obsolete switch causes Perl to dump core after compiling your
program.  You can then in theory take this core dump and turn it
into an executable file by using the B<undump> program (not supplied).
This speeds startup at the expense of some disk space (which you
can minimize by stripping the executable).  (Still, a "hello world"
executable comes out to about 200K on my machine.)  If you want to
execute a portion of your program before dumping, use the dump()
operator instead.  Note: availability of B<undump> is platform
specific and may not be available for a specific port of Perl.

This switch has been superseded in favor of the new Perl code
generator backends to the compiler.  See L<B> and L<B::Bytecode>
for details.

=item B<-U>

allows Perl to do unsafe operations.  Currently the only "unsafe"
operations are the unlinking of directories while running as superuser,
and running setuid programs with fatal taint checks turned into
warnings.  Note that the B<-w> switch (or the C<$^W> variable) must
be used along with this option to actually I<generate> the
taint-check warnings.

=item B<-v>

prints the version and patchlevel of your perl executable.

=item B<-V>

prints summary of the major perl configuration values and the current
values of @INC.

=item B<-V:>I<configvar>

Prints to STDOUT the value of the named configuration variable(s),
with multiples when your configvar argument looks like a regex (has
non-letters).  For example:

    $ perl -V:libc
	libc='/lib/libc-2.2.4.so';
    $ perl -V:lib.
	libs='-lnsl -lgdbm -ldb -ldl -lm -lcrypt -lutil -lc';
	libc='/lib/libc-2.2.4.so';
    $ perl -V:lib.*
	libpth='/usr/local/lib /lib /usr/lib';
	libs='-lnsl -lgdbm -ldb -ldl -lm -lcrypt -lutil -lc';
	lib_ext='.a';
	libc='/lib/libc-2.2.4.so';
	libperl='libperl.a';
	....

Additionally, extra colons can be used to control formatting.  A
trailing colon suppresses the linefeed and terminator ';', allowing
you to embed queries into shell commands.  (mnemonic: PATH separator
':'.)

    $ echo "compression-vars: " `perl -V:z.*: ` " are here !"
    compression-vars:  zcat='' zip='zip'  are here !

A leading colon removes the 'name=' part of the response, this allows
you to map to the name you need.  (mnemonic: empty label)

    $ echo "goodvfork="`./perl -Ilib -V::usevfork`
    goodvfork=false;

Leading and trailing colons can be used together if you need
positional parameter values without the names.  Note that in the case
below, the PERL_API params are returned in alphabetical order.

    $ echo building_on `perl -V::osname: -V::PERL_API_.*:` now
    building_on 'linux' '5' '1' '9' now

=item B<-w>

prints warnings about dubious constructs, such as variable names
that are mentioned only once and scalar variables that are used
before being set, redefined subroutines, references to undefined
filehandles or filehandles opened read-only that you are attempting
to write on, values used as a number that doesn't look like numbers,
using an array as though it were a scalar, if your subroutines
recurse more than 100 deep, and innumerable other things.

This switch really just enables the internal C<$^W> variable.  You
can disable or promote into fatal errors specific warnings using
C<__WARN__> hooks, as described in L<perlvar> and L<perlfunc/warn>.
See also L<perldiag> and L<perltrap>.  A new, fine-grained warning
facility is also available if you want to manipulate entire classes
of warnings; see L<warnings> or L<perllexwarn>.

=item B<-W>

Enables all warnings regardless of C<no warnings> or C<$^W>.
See L<perllexwarn>.

=item B<-X>

Disables all warnings regardless of C<use warnings> or C<$^W>.
See L<perllexwarn>.

=item B<-x>

=item B<-x> I<directory>

tells Perl that the program is embedded in a larger chunk of unrelated
ASCII text, such as in a mail message.  Leading garbage will be
discarded until the first line that starts with #! and contains the
string "perl".  Any meaningful switches on that line will be applied.
If a directory name is specified, Perl will switch to that directory
before running the program.  The B<-x> switch controls only the
disposal of leading garbage.  The program must be terminated with
C<__END__> if there is trailing garbage to be ignored (the program
can process any or all of the trailing garbage via the DATA filehandle
if desired).

=back

=head1 ENVIRONMENT

=over 12

=item HOME

Used if chdir has no argument.

=item LOGDIR

Used if chdir has no argument and HOME is not set.

=item PATH

Used in executing subprocesses, and in finding the program if B<-S> is
used.

=item PERL5LIB

A list of directories in which to look for Perl library
files before looking in the standard library and the current
directory.  Any architecture-specific directories under the specified
locations are automatically included if they exist.  If PERL5LIB is not
defined, PERLLIB is used.  Directories are separated (like in PATH) by
a colon on unixish platforms and by a semicolon on Windows (the proper
path separator being given by the command C<perl -V:path_sep>).

When running taint checks (either because the program was running setuid
or setgid, or the B<-T> switch was used), neither variable is used.
The program should instead say:

    use lib "/my/directory";

=item PERL5OPT

Command-line options (switches).  Switches in this variable are taken
as if they were on every Perl command line.  Only the B<-[DIMUdmtw]>
switches are allowed.  When running taint checks (because the program
was running setuid or setgid, or the B<-T> switch was used), this
variable is ignored.  If PERL5OPT begins with B<-T>, tainting will be
enabled, and any subsequent options ignored.

=item PERLIO

A space (or colon) separated list of PerlIO layers. If perl is built
to use PerlIO system for IO (the default) these layers effect perl's IO.

It is conventional to start layer names with a colon e.g. C<:perlio> to
emphasise their similarity to variable "attributes". But the code that parses
layer specification strings (which is also used to decode the PERLIO
environment variable) treats the colon as a separator.

An unset or empty PERLIO is equivalent to C<:stdio>.

The list becomes the default for I<all> perl's IO. Consequently only built-in
layers can appear in this list, as external layers (such as :encoding()) need
IO in  order to load them!. See L<"open pragma"|open> for how to add external
encodings as defaults.

The layers that it makes sense to include in the PERLIO environment
variable are briefly summarised below. For more details see L<PerlIO>.

=over 8

=item :bytes

A pseudolayer that turns I<off> the C<:utf8> flag for the layer below.
Unlikely to be useful on its own in the global PERLIO environment variable.
You perhaps were thinking of C<:crlf:bytes> or C<:perlio:bytes>.

=item :crlf

A layer which does CRLF to "\n" translation distinguishing "text" and
"binary" files in the manner of MS-DOS and similar operating systems.
(It currently does I<not> mimic MS-DOS as far as treating of Control-Z
as being an end-of-file marker.)

=item :mmap

A layer which implements "reading" of files by using C<mmap()> to
make (whole) file appear in the process's address space, and then
using that as PerlIO's "buffer".

=item :perlio

This is a re-implementation of "stdio-like" buffering written as a
PerlIO "layer".  As such it will call whatever layer is below it for
its operations (typically C<:unix>).

=item :pop

An experimental pseudolayer that removes the topmost layer.
Use with the same care as is reserved for nitroglycerin.

=item :raw

A pseudolayer that manipulates other layers.  Applying the C<:raw>
layer is equivalent to calling C<binmode($fh)>.  It makes the stream
pass each byte as-is without any translation.  In particular CRLF
translation, and/or :utf8 intuited from locale are disabled.

Unlike in the earlier versions of Perl C<:raw> is I<not>
just the inverse of C<:crlf> - other layers which would affect the
binary nature of the stream are also removed or disabled.

=item :stdio

This layer provides PerlIO interface by wrapping system's ANSI C "stdio"
library calls. The layer provides both buffering and IO.
Note that C<:stdio> layer does I<not> do CRLF translation even if that
is platforms normal behaviour. You will need a C<:crlf> layer above it
to do that.

=item :unix

Low level layer which calls C<read>, C<write> and C<lseek> etc.

=item :utf8

A pseudolayer that turns on a flag on the layer below to tell perl
that output should be in utf8 and that input should be regarded as
already in utf8 form.  May be useful in PERLIO environment
variable to make UTF-8 the default. (To turn off that behaviour
use C<:bytes> layer.)

=item :win32

On Win32 platforms this I<experimental> layer uses native "handle" IO
rather than unix-like numeric file descriptor layer. Known to be
buggy in this release.

=back

On all platforms the default set of layers should give acceptable results.

For UNIX platforms that will equivalent of "unix perlio" or "stdio".
Configure is setup to prefer "stdio" implementation if system's library
provides for fast access to the buffer, otherwise it uses the "unix perlio"
implementation.

On Win32 the default in this release is "unix crlf". Win32's "stdio"
has a number of bugs/mis-features for perl IO which are somewhat
C compiler vendor/version dependent. Using our own C<crlf> layer as
the buffer avoids those issues and makes things more uniform.
The C<crlf> layer provides CRLF to/from "\n" conversion as well as
buffering.

This release uses C<unix> as the bottom layer on Win32 and so still uses C
compiler's numeric file descriptor routines. There is an experimental native
C<win32> layer which is expected to be enhanced and should eventually be
the default under Win32.

=item PERLIO_DEBUG

If set to the name of a file or device then certain operations of PerlIO
sub-system will be logged to that file (opened as append). Typical uses
are UNIX:

   PERLIO_DEBUG=/dev/tty perl script ...

and Win32 approximate equivalent:

   set PERLIO_DEBUG=CON
   perl script ...

This functionality is disabled for setuid scripts and for scripts run
with B<-T>.

=item PERLLIB

A list of directories in which to look for Perl library
files before looking in the standard library and the current directory.
If PERL5LIB is defined, PERLLIB is not used.

=item PERL5DB

The command used to load the debugger code.  The default is:

	BEGIN { require 'perl5db.pl' }

=item PERL5DB_THREADED

If set to a true value, indicates to the debugger that the code being
debugged uses threads.

=item PERL5SHELL (specific to the Win32 port)

May be set to an alternative shell that perl must use internally for
executing "backtick" commands or system().  Default is C<cmd.exe /x/d/c>
on WindowsNT and C<command.com /c> on Windows95.  The value is considered
to be space-separated.  Precede any character that needs to be protected
(like a space or backslash) with a backslash.

Note that Perl doesn't use COMSPEC for this purpose because
COMSPEC has a high degree of variability among users, leading to
portability concerns.  Besides, perl can use a shell that may not be
fit for interactive use, and setting COMSPEC to such a shell may
interfere with the proper functioning of other programs (which usually
look in COMSPEC to find a shell fit for interactive use).

=item PERL_ALLOW_NON_IFS_LSP (specific to the Win32 port)

Set to 1 to allow the use of non-IFS compatible LSP's.
Perl normally searches for an IFS-compatible LSP because this is required
for its emulation of Windows sockets as real filehandles.  However, this may
cause problems if you have a firewall such as McAfee Guardian which requires
all applications to use its LSP which is not IFS-compatible, because clearly
Perl will normally avoid using such an LSP.
Setting this environment variable to 1 means that Perl will simply use the
first suitable LSP enumerated in the catalog, which keeps McAfee Guardian
happy (and in that particular case Perl still works too because McAfee
Guardian's LSP actually plays some other games which allow applications
requiring IFS compatibility to work).

=item PERL_DEBUG_MSTATS

Relevant only if perl is compiled with the malloc included with the perl
distribution (that is, if C<perl -V:d_mymalloc> is 'define').
If set, this causes memory statistics to be dumped after execution.  If set
to an integer greater than one, also causes memory statistics to be dumped
after compilation.

=item PERL_DESTRUCT_LEVEL

Relevant only if your perl executable was built with B<-DDEBUGGING>,
this controls the behavior of global destruction of objects and other
references.  See L<perlhack/PERL_DESTRUCT_LEVEL> for more information.

=item PERL_DL_NONLAZY

Set to one to have perl resolve B<all> undefined symbols when it loads
a dynamic library.  The default behaviour is to resolve symbols when
they are used.  Setting this variable is useful during testing of
extensions as it ensures that you get an error on misspelled function
names even if the test suite doesn't call it.

=item PERL_ENCODING

If using the C<encoding> pragma without an explicit encoding name, the
PERL_ENCODING environment variable is consulted for an encoding name.

=item PERL_HASH_SEED

(Since Perl 5.8.1.)  Used to randomise Perl's internal hash function.
To emulate the pre-5.8.1 behaviour, set to an integer (zero means
exactly the same order as 5.8.0).  "Pre-5.8.1" means, among other
things, that hash keys will be ordered the same between different runs
of Perl.

The default behaviour is to randomise unless the PERL_HASH_SEED is set.
If Perl has been compiled with C<-DUSE_HASH_SEED_EXPLICIT>, the default
behaviour is B<not> to randomise unless the PERL_HASH_SEED is set.

If PERL_HASH_SEED is unset or set to a non-numeric string, Perl uses
the pseudorandom seed supplied by the operating system and libraries.
This means that each different run of Perl will have a different
ordering of the results of keys(), values(), and each().

B<Please note that the hash seed is sensitive information>. Hashes are
randomized to protect against local and remote attacks against Perl
code. By manually setting a seed this protection may be partially or
completely lost.

See L<perlsec/"Algorithmic Complexity Attacks"> and
L</PERL_HASH_SEED_DEBUG> for more information.

=item PERL_HASH_SEED_DEBUG

(Since Perl 5.8.1.)  Set to one to display (to STDERR) the value of
the hash seed at the beginning of execution.  This, combined with
L</PERL_HASH_SEED> is intended to aid in debugging nondeterministic
behavior caused by hash randomization.

B<Note that the hash seed is sensitive information>: by knowing it one
can craft a denial-of-service attack against Perl code, even remotely,
see L<perlsec/"Algorithmic Complexity Attacks"> for more information.
B<Do not disclose the hash seed> to people who don't need to know it.
See also hash_seed() of L<Hash::Util>.

=item PERL_ROOT (specific to the VMS port)

A translation concealed rooted logical name that contains perl and the
logical device for the @INC path on VMS only.  Other logical names that
affect perl on VMS include PERLSHR, PERL_ENV_TABLES, and
SYS$TIMEZONE_DIFFERENTIAL but are optional and discussed further in
L<perlvms> and in F<README.vms> in the Perl source distribution.

=item PERL_SIGNALS

In Perls 5.8.1 and later.  If set to C<unsafe> the pre-Perl-5.8.0
signals behaviour (immediate but unsafe) is restored.  If set to
C<safe> the safe (or deferred) signals are used.
See L<perlipc/"Deferred Signals (Safe Signals)">.

=item PERL_UNICODE

Equivalent to the B<-C> command-line switch.  Note that this is not
a boolean variable-- setting this to C<"1"> is not the right way to
"enable Unicode" (whatever that would mean).  You can use C<"0"> to
"disable Unicode", though (or alternatively unset PERL_UNICODE in
your shell before starting Perl).  See the description of the C<-C>
switch for more information.

=item SYS$LOGIN (specific to the VMS port)

Used if chdir has no argument and HOME and LOGDIR are not set.

=back

Perl also has environment variables that control how Perl handles data
specific to particular natural languages.  See L<perllocale>.

Apart from these, Perl uses no other environment variables, except
to make them available to the program being executed, and to child
processes.  However, programs running setuid would do well to execute
the following lines before doing anything else, just to keep people
honest:

    $ENV{PATH}  = '/bin:/usr/bin';    # or whatever you need
    $ENV{SHELL} = '/bin/sh' if exists $ENV{SHELL};
    delete @ENV{qw(IFS CDPATH ENV BASH_ENV)};