This is a live mirror of the Perl 5 development currently hosted at https://github.com/perl/perl5
perl 5.000
[perl5.git] / pod / perlrun.pod
CommitLineData
a0d0e21e
LW
1=head1 NAME
2
3perlrun - how to execute the Perl interpreter
4
5=head1 SYNOPSIS
6
7B<perl> [switches] filename args
8
9=head1 DESCRIPTION
10
11Upon startup, Perl looks for your script in one of the following
12places:
13
14=over 4
15
16=item 1.
17
18Specified line by line via B<-e> switches on the command line.
19
20=item 2.
21
22Contained in the file specified by the first filename on the command line.
23(Note that systems supporting the #! notation invoke interpreters this way.)
24
25=item 3.
26
27Passed in implicitly via standard input. This only works if there are
28no filename arguments--to pass arguments to a STDIN script you
29must explicitly specify a "-" for the script name.
30
31=back
32
33With methods 2 and 3, Perl starts parsing the input file from the
34beginning, unless you've specified a B<-x> switch, in which case it
35scans for the first line starting with #! and containing the word
36"perl", and starts there instead. This is useful for running a script
37embedded in a larger message. (In this case you would indicate the end
38of the script using the __END__ token.)
39
40As of Perl 5, the #! line is always examined for switches as the line is
41being parsed. Thus, if you're on a machine that only allows one argument
42with the #! line, or worse, doesn't even recognize the #! line, you still
43can get consistent switch behavior regardless of how Perl was invoked,
44even if B<-x> was used to find the beginning of the script.
45
46Because many operating systems silently chop off kernel interpretation of
47the #! line after 32 characters, some switches may be passed in on the
48command line, and some may not; you could even get a "-" without its
49letter, if you're not careful. You probably want to make sure that all
50your switches fall either before or after that 32 character boundary.
51Most switches don't actually care if they're processed redundantly, but
52getting a - instead of a complete switch could cause Perl to try to
53execute standard input instead of your script. And a partial B<-I> switch
54could also cause odd results.
55
56Parsing of the #! switches starts wherever "perl" is mentioned in the line.
57The sequences "-*" and "- " are specifically ignored so that you could,
58if you were so inclined, say
59
60 #!/bin/sh -- # -*- perl -*- -p
61 eval 'exec perl $0 -S ${1+"$@"}'
62 if 0;
63
64to let Perl see the B<-p> switch.
65
66If the #! line does not contain the word "perl", the program named after
67the #! is executed instead of the Perl interpreter. This is slightly
68bizarre, but it helps people on machines that don't do #!, because they
69can tell a program that their SHELL is /usr/bin/perl, and Perl will then
70dispatch the program to the correct interpreter for them.
71
72After locating your script, Perl compiles the entire script to an
73internal form. If there are any compilation errors, execution of the
74script is not attempted. (This is unlike the typical shell script,
75which might run partway through before finding a syntax error.)
76
77If the script is syntactically correct, it is executed. If the script
78runs off the end without hitting an exit() or die() operator, an implicit
79C<exit(0)> is provided to indicate successful completion.
80
81=head2 Switches
82
83A single-character switch may be combined with the following switch, if
84any.
85
86 #!/usr/bin/perl -spi.bak # same as -s -p -i.bak
87
88Switches include:
89
90=over 5
91
92=item B<-0>I<digits>
93
94specifies the record separator (C<$/>) as an octal number. If there are
95no digits, the null character is the separator. Other switches may
96precede or follow the digits. For example, if you have a version of
97B<find> which can print filenames terminated by the null character, you
98can say this:
99
100 find . -name '*.bak' -print0 | perl -n0e unlink
101
102The special value 00 will cause Perl to slurp files in paragraph mode.
103The value 0777 will cause Perl to slurp files whole since there is no
104legal character with that value.
105
106=item B<-a>
107
108turns on autosplit mode when used with a B<-n> or B<-p>. An implicit
109split command to the @F array is done as the first thing inside the
110implicit while loop produced by the B<-n> or B<-p>.
111
112 perl -ane 'print pop(@F), "\n";'
113
114is equivalent to
115
116 while (<>) {
117 @F = split(' ');
118 print pop(@F), "\n";
119 }
120
121An alternate delimiter may be specified using B<-F>.
122
123=item B<-c>
124
125causes Perl to check the syntax of the script and then exit without
126executing it.
127
128=item B<-d>
129
130runs the script under the Perl debugger. See L<perldebug>.
131
132=item B<-D>I<number>
133
134=item B<-D>I<list>
135
136sets debugging flags. To watch how it executes your script, use
137B<-D14>. (This only works if debugging is compiled into your
138Perl.) Another nice value is B<-D1024>, which lists your compiled
139syntax tree. And B<-D512> displays compiled regular expressions. As an
140alternative specify a list of letters instead of numbers (e.g. B<-D14> is
141equivalent to B<-Dtls>):
142
143 1 p Tokenizing and Parsing
144 2 s Stack Snapshots
145 4 l Label Stack Processing
146 8 t Trace Execution
147 16 o Operator Node Construction
148 32 c String/Numeric Conversions
149 64 P Print Preprocessor Command for -P
150 128 m Memory Allocation
151 256 f Format Processing
152 512 r Regular Expression Parsing
153 1024 x Syntax Tree Dump
154 2048 u Tainting Checks
155 4096 L Memory Leaks (not supported anymore)
156 8192 H Hash Dump -- usurps values()
157 16384 X Scratchpad Allocation
158 32768 D Cleaning Up
159
160=item B<-e> I<commandline>
161
162may be used to enter one line of script.
163If B<-e> is given, Perl
164will not look for a script filename in the argument list.
165Multiple B<-e> commands may
166be given to build up a multi-line script.
167Make sure to use semicolons where you would in a normal program.
168
169=item B<-F>I<regexp>
170
171specifies a regular expression to split on if B<-a> is also in effect.
172If regexp has C<//> around it, the slashes will be ignored.
173
174=item B<-i>I<extension>
175
176specifies that files processed by the C<E<lt>E<gt>> construct are to be edited
177in-place. It does this by renaming the input file, opening the output
178file by the original name, and selecting that output file as the default
179for print() statements. The extension, if supplied, is added to the name
180of the old file to make a backup copy. If no extension is supplied, no
181backup is made. From the shell, saying
182
183 $ perl -p -i.bak -e "s/foo/bar/; ... "
184
185is the same as using the script:
186
187 #!/usr/bin/perl -pi.bak
188 s/foo/bar/;
189
190which is equivalent to
191
192 #!/usr/bin/perl
193 while (<>) {
194 if ($ARGV ne $oldargv) {
195 rename($ARGV, $ARGV . '.bak');
196 open(ARGVOUT, ">$ARGV");
197 select(ARGVOUT);
198 $oldargv = $ARGV;
199 }
200 s/foo/bar/;
201 }
202 continue {
203 print; # this prints to original filename
204 }
205 select(STDOUT);
206
207except that the B<-i> form doesn't need to compare $ARGV to $oldargv to
208know when the filename has changed. It does, however, use ARGVOUT for
209the selected filehandle. Note that STDOUT is restored as the
210default output filehandle after the loop.
211
212You can use C<eof> without parenthesis to locate the end of each input file,
213in case you want to append to each file, or reset line numbering (see
214example in L<perlfunc/eof>).
215
216=item B<-I>I<directory>
217
218may be used in conjunction with B<-P> to tell the C preprocessor where
219to look for include files. By default /usr/include and /usr/lib/perl
220are searched.
221
222=item B<-l>I<octnum>
223
224enables automatic line-ending processing. It has two effects: first,
225it automatically chomps the line terminator when used with B<-n> or
226B<-p>, and second, it assigns "C<$\>" to have the value of I<octnum> so that
227any print statements will have that line terminator added back on. If
228I<octnum> is omitted, sets "C<$\>" to the current value of "C<$/>". For
229instance, to trim lines to 80 columns:
230
231 perl -lpe 'substr($_, 80) = ""'
232
233Note that the assignment C<$\ = $/> is done when the switch is processed,
234so the input record separator can be different than the output record
235separator if the B<-l> switch is followed by a B<-0> switch:
236
237 gnufind / -print0 | perl -ln0e 'print "found $_" if -p'
238
239This sets $\ to newline and then sets $/ to the null character.
240
241=item B<-n>
242
243causes Perl to assume the following loop around your script, which
244makes it iterate over filename arguments somewhat like B<sed -n> or
245B<awk>:
246
247 while (<>) {
248 ... # your script goes here
249 }
250
251Note that the lines are not printed by default. See B<-p> to have
252lines printed. Here is an efficient way to delete all files older than
253a week:
254
255 find . -mtime +7 -print | perl -nle 'unlink;'
256
257This is faster than using the C<-exec> switch of B<find> because you don't
258have to start a process on every filename found.
259
260C<BEGIN> and C<END> blocks may be used to capture control before or after
261the implicit loop, just as in B<awk>.
262
263=item B<-p>
264
265causes Perl to assume the following loop around your script, which
266makes it iterate over filename arguments somewhat like B<sed>:
267
268
269 while (<>) {
270 ... # your script goes here
271 } continue {
272 print;
273 }
274
275Note that the lines are printed automatically. To suppress printing
276use the B<-n> switch. A B<-p> overrides a B<-n> switch.
277
278C<BEGIN> and C<END> blocks may be used to capture control before or after
279the implicit loop, just as in awk.
280
281=item B<-P>
282
283causes your script to be run through the C preprocessor before
284compilation by Perl. (Since both comments and cpp directives begin
285with the # character, you should avoid starting comments with any words
286recognized by the C preprocessor such as "if", "else" or "define".)
287
288=item B<-s>
289
290enables some rudimentary switch parsing for switches on the command
291line after the script name but before any filename arguments (or before
292a B<-->). Any switch found there is removed from @ARGV and sets the
293corresponding variable in the Perl script. The following script
294prints "true" if and only if the script is invoked with a B<-xyz> switch.
295
296 #!/usr/bin/perl -s
297 if ($xyz) { print "true\n"; }
298
299=item B<-S>
300
301makes Perl use the PATH environment variable to search for the
302script (unless the name of the script starts with a slash). Typically
303this is used to emulate #! startup on machines that don't support #!,
304in the following manner:
305
306 #!/usr/bin/perl
307 eval "exec /usr/bin/perl -S $0 $*"
308 if $running_under_some_shell;
309
310The system ignores the first line and feeds the script to /bin/sh,
311which proceeds to try to execute the Perl script as a shell script.
312The shell executes the second line as a normal shell command, and thus
313starts up the Perl interpreter. On some systems $0 doesn't always
314contain the full pathname, so the B<-S> tells Perl to search for the
315script if necessary. After Perl locates the script, it parses the
316lines and ignores them because the variable $running_under_some_shell
317is never true. A better construct than C<$*> would be C<${1+"$@"}>, which
318handles embedded spaces and such in the filenames, but doesn't work if
319the script is being interpreted by csh. In order to start up sh rather
320than csh, some systems may have to replace the #! line with a line
321containing just a colon, which will be politely ignored by Perl. Other
322systems can't control that, and need a totally devious construct that
323will work under any of csh, sh or Perl, such as the following:
324
325 eval '(exit $?0)' && eval 'exec /usr/bin/perl -S $0 ${1+"$@"}'
326 & eval 'exec /usr/bin/perl -S $0 $argv:q'
327 if 0;
328
329=item B<-T>
330
331forces "taint" checks to be turned on. Ordinarily these checks are
332done only when running setuid or setgid. See L<perlsec>.
333
334=item B<-u>
335
336causes Perl to dump core after compiling your script. You can then
337take this core dump and turn it into an executable file by using the
338B<undump> program (not supplied). This speeds startup at the expense of
339some disk space (which you can minimize by stripping the executable).
340(Still, a "hello world" executable comes out to about 200K on my
341machine.) If you want to execute a portion of your script before dumping,
342use the dump() operator instead. Note: availability of B<undump> is
343platform specific and may not be available for a specific port of
344Perl.
345
346=item B<-U>
347
348allows Perl to do unsafe operations. Currently the only "unsafe"
349operations are the unlinking of directories while running as superuser,
350and running setuid programs with fatal taint checks turned into
351warnings.
352
353=item B<-v>
354
355prints the version and patchlevel of your Perl executable.
356
357=item B<-w>
358
359prints warnings about identifiers that are mentioned only once, and
360scalar variables that are used before being set. Also warns about
361redefined subroutines, and references to undefined filehandles or
362filehandles opened readonly that you are attempting to write on. Also
363warns you if you use values as a number that doesn't look like numbers, using
364a an array as though it were a scalar, if
365your subroutines recurse more than 100 deep, and innumeriable other things.
366See L<perldiag> and L<perltrap>.
367
368=item B<-x> I<directory>
369
370tells Perl that the script is embedded in a message. Leading
371garbage will be discarded until the first line that starts with #! and
372contains the string "perl". Any meaningful switches on that line will
373be applied (but only one group of switches, as with normal #!
374processing). If a directory name is specified, Perl will switch to
375that directory before running the script. The B<-x> switch only
376controls the the disposal of leading garbage. The script must be
377terminated with C<__END__> if there is trailing garbage to be ignored (the
378script can process any or all of the trailing garbage via the DATA
379filehandle if desired).
380
381
382=back