Commit | Line | Data |
a0d0e21e |
1 | =head1 NAME |
2 | |
3 | perlrun - how to execute the Perl interpreter |
4 | |
5 | =head1 SYNOPSIS |
6 | |
7 | B<perl> [switches] filename args |
8 | |
9 | =head1 DESCRIPTION |
10 | |
11 | Upon startup, Perl looks for your script in one of the following |
12 | places: |
13 | |
14 | =over 4 |
15 | |
16 | =item 1. |
17 | |
18 | Specified line by line via B<-e> switches on the command line. |
19 | |
20 | =item 2. |
21 | |
22 | Contained 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 | |
27 | Passed in implicitly via standard input. This only works if there are |
28 | no filename arguments--to pass arguments to a STDIN script you |
29 | must explicitly specify a "-" for the script name. |
30 | |
31 | =back |
32 | |
33 | With methods 2 and 3, Perl starts parsing the input file from the |
34 | beginning, unless you've specified a B<-x> switch, in which case it |
35 | scans for the first line starting with #! and containing the word |
36 | "perl", and starts there instead. This is useful for running a script |
37 | embedded in a larger message. (In this case you would indicate the end |
38 | of the script using the __END__ token.) |
39 | |
40 | As of Perl 5, the #! line is always examined for switches as the line is |
41 | being parsed. Thus, if you're on a machine that only allows one argument |
42 | with the #! line, or worse, doesn't even recognize the #! line, you still |
43 | can get consistent switch behavior regardless of how Perl was invoked, |
44 | even if B<-x> was used to find the beginning of the script. |
45 | |
46 | Because many operating systems silently chop off kernel interpretation of |
47 | the #! line after 32 characters, some switches may be passed in on the |
48 | command line, and some may not; you could even get a "-" without its |
49 | letter, if you're not careful. You probably want to make sure that all |
50 | your switches fall either before or after that 32 character boundary. |
51 | Most switches don't actually care if they're processed redundantly, but |
52 | getting a - instead of a complete switch could cause Perl to try to |
53 | execute standard input instead of your script. And a partial B<-I> switch |
54 | could also cause odd results. |
55 | |
56 | Parsing of the #! switches starts wherever "perl" is mentioned in the line. |
57 | The sequences "-*" and "- " are specifically ignored so that you could, |
58 | if you were so inclined, say |
59 | |
60 | #!/bin/sh -- # -*- perl -*- -p |
61 | eval 'exec perl $0 -S ${1+"$@"}' |
62 | if 0; |
63 | |
64 | to let Perl see the B<-p> switch. |
65 | |
66 | If the #! line does not contain the word "perl", the program named after |
67 | the #! is executed instead of the Perl interpreter. This is slightly |
68 | bizarre, but it helps people on machines that don't do #!, because they |
69 | can tell a program that their SHELL is /usr/bin/perl, and Perl will then |
70 | dispatch the program to the correct interpreter for them. |
71 | |
72 | After locating your script, Perl compiles the entire script to an |
73 | internal form. If there are any compilation errors, execution of the |
74 | script is not attempted. (This is unlike the typical shell script, |
75 | which might run partway through before finding a syntax error.) |
76 | |
77 | If the script is syntactically correct, it is executed. If the script |
78 | runs off the end without hitting an exit() or die() operator, an implicit |
79 | C<exit(0)> is provided to indicate successful completion. |
80 | |
81 | =head2 Switches |
82 | |
83 | A single-character switch may be combined with the following switch, if |
84 | any. |
85 | |
86 | #!/usr/bin/perl -spi.bak # same as -s -p -i.bak |
87 | |
88 | Switches include: |
89 | |
90 | =over 5 |
91 | |
92 | =item B<-0>I<digits> |
93 | |
94 | specifies the record separator (C<$/>) as an octal number. If there are |
95 | no digits, the null character is the separator. Other switches may |
96 | precede or follow the digits. For example, if you have a version of |
97 | B<find> which can print filenames terminated by the null character, you |
98 | can say this: |
99 | |
100 | find . -name '*.bak' -print0 | perl -n0e unlink |
101 | |
102 | The special value 00 will cause Perl to slurp files in paragraph mode. |
103 | The value 0777 will cause Perl to slurp files whole since there is no |
104 | legal character with that value. |
105 | |
106 | =item B<-a> |
107 | |
108 | turns on autosplit mode when used with a B<-n> or B<-p>. An implicit |
109 | split command to the @F array is done as the first thing inside the |
110 | implicit while loop produced by the B<-n> or B<-p>. |
111 | |
112 | perl -ane 'print pop(@F), "\n";' |
113 | |
114 | is equivalent to |
115 | |
116 | while (<>) { |
117 | @F = split(' '); |
118 | print pop(@F), "\n"; |
119 | } |
120 | |
121 | An alternate delimiter may be specified using B<-F>. |
122 | |
123 | =item B<-c> |
124 | |
125 | causes Perl to check the syntax of the script and then exit without |
748a9306 |
126 | executing it. Actually, it will execute C<BEGIN> and C<use> blocks, |
127 | since these are considered part of the compilation. |
a0d0e21e |
128 | |
129 | =item B<-d> |
130 | |
131 | runs the script under the Perl debugger. See L<perldebug>. |
132 | |
133 | =item B<-D>I<number> |
134 | |
135 | =item B<-D>I<list> |
136 | |
137 | sets debugging flags. To watch how it executes your script, use |
138 | B<-D14>. (This only works if debugging is compiled into your |
139 | Perl.) Another nice value is B<-D1024>, which lists your compiled |
140 | syntax tree. And B<-D512> displays compiled regular expressions. As an |
141 | alternative specify a list of letters instead of numbers (e.g. B<-D14> is |
142 | equivalent to B<-Dtls>): |
143 | |
144 | 1 p Tokenizing and Parsing |
145 | 2 s Stack Snapshots |
146 | 4 l Label Stack Processing |
147 | 8 t Trace Execution |
148 | 16 o Operator Node Construction |
149 | 32 c String/Numeric Conversions |
150 | 64 P Print Preprocessor Command for -P |
151 | 128 m Memory Allocation |
152 | 256 f Format Processing |
153 | 512 r Regular Expression Parsing |
154 | 1024 x Syntax Tree Dump |
155 | 2048 u Tainting Checks |
156 | 4096 L Memory Leaks (not supported anymore) |
157 | 8192 H Hash Dump -- usurps values() |
158 | 16384 X Scratchpad Allocation |
159 | 32768 D Cleaning Up |
160 | |
161 | =item B<-e> I<commandline> |
162 | |
163 | may be used to enter one line of script. |
164 | If B<-e> is given, Perl |
165 | will not look for a script filename in the argument list. |
166 | Multiple B<-e> commands may |
167 | be given to build up a multi-line script. |
168 | Make sure to use semicolons where you would in a normal program. |
169 | |
170 | =item B<-F>I<regexp> |
171 | |
172 | specifies a regular expression to split on if B<-a> is also in effect. |
173 | If regexp has C<//> around it, the slashes will be ignored. |
174 | |
175 | =item B<-i>I<extension> |
176 | |
177 | specifies that files processed by the C<E<lt>E<gt>> construct are to be edited |
178 | in-place. It does this by renaming the input file, opening the output |
179 | file by the original name, and selecting that output file as the default |
180 | for print() statements. The extension, if supplied, is added to the name |
181 | of the old file to make a backup copy. If no extension is supplied, no |
182 | backup is made. From the shell, saying |
183 | |
184 | $ perl -p -i.bak -e "s/foo/bar/; ... " |
185 | |
186 | is the same as using the script: |
187 | |
188 | #!/usr/bin/perl -pi.bak |
189 | s/foo/bar/; |
190 | |
191 | which is equivalent to |
192 | |
193 | #!/usr/bin/perl |
194 | while (<>) { |
195 | if ($ARGV ne $oldargv) { |
196 | rename($ARGV, $ARGV . '.bak'); |
197 | open(ARGVOUT, ">$ARGV"); |
198 | select(ARGVOUT); |
199 | $oldargv = $ARGV; |
200 | } |
201 | s/foo/bar/; |
202 | } |
203 | continue { |
204 | print; # this prints to original filename |
205 | } |
206 | select(STDOUT); |
207 | |
208 | except that the B<-i> form doesn't need to compare $ARGV to $oldargv to |
209 | know when the filename has changed. It does, however, use ARGVOUT for |
210 | the selected filehandle. Note that STDOUT is restored as the |
211 | default output filehandle after the loop. |
212 | |
213 | You can use C<eof> without parenthesis to locate the end of each input file, |
214 | in case you want to append to each file, or reset line numbering (see |
215 | example in L<perlfunc/eof>). |
216 | |
217 | =item B<-I>I<directory> |
218 | |
219 | may be used in conjunction with B<-P> to tell the C preprocessor where |
220 | to look for include files. By default /usr/include and /usr/lib/perl |
221 | are searched. |
222 | |
223 | =item B<-l>I<octnum> |
224 | |
225 | enables automatic line-ending processing. It has two effects: first, |
226 | it automatically chomps the line terminator when used with B<-n> or |
227 | B<-p>, and second, it assigns "C<$\>" to have the value of I<octnum> so that |
228 | any print statements will have that line terminator added back on. If |
229 | I<octnum> is omitted, sets "C<$\>" to the current value of "C<$/>". For |
230 | instance, to trim lines to 80 columns: |
231 | |
232 | perl -lpe 'substr($_, 80) = ""' |
233 | |
234 | Note that the assignment C<$\ = $/> is done when the switch is processed, |
235 | so the input record separator can be different than the output record |
236 | separator if the B<-l> switch is followed by a B<-0> switch: |
237 | |
238 | gnufind / -print0 | perl -ln0e 'print "found $_" if -p' |
239 | |
240 | This sets $\ to newline and then sets $/ to the null character. |
241 | |
242 | =item B<-n> |
243 | |
244 | causes Perl to assume the following loop around your script, which |
245 | makes it iterate over filename arguments somewhat like B<sed -n> or |
246 | B<awk>: |
247 | |
248 | while (<>) { |
249 | ... # your script goes here |
250 | } |
251 | |
252 | Note that the lines are not printed by default. See B<-p> to have |
253 | lines printed. Here is an efficient way to delete all files older than |
254 | a week: |
255 | |
256 | find . -mtime +7 -print | perl -nle 'unlink;' |
257 | |
258 | This is faster than using the C<-exec> switch of B<find> because you don't |
259 | have to start a process on every filename found. |
260 | |
261 | C<BEGIN> and C<END> blocks may be used to capture control before or after |
262 | the implicit loop, just as in B<awk>. |
263 | |
264 | =item B<-p> |
265 | |
266 | causes Perl to assume the following loop around your script, which |
267 | makes it iterate over filename arguments somewhat like B<sed>: |
268 | |
269 | |
270 | while (<>) { |
271 | ... # your script goes here |
272 | } continue { |
273 | print; |
274 | } |
275 | |
276 | Note that the lines are printed automatically. To suppress printing |
277 | use the B<-n> switch. A B<-p> overrides a B<-n> switch. |
278 | |
279 | C<BEGIN> and C<END> blocks may be used to capture control before or after |
280 | the implicit loop, just as in awk. |
281 | |
282 | =item B<-P> |
283 | |
284 | causes your script to be run through the C preprocessor before |
285 | compilation by Perl. (Since both comments and cpp directives begin |
286 | with the # character, you should avoid starting comments with any words |
287 | recognized by the C preprocessor such as "if", "else" or "define".) |
288 | |
289 | =item B<-s> |
290 | |
291 | enables some rudimentary switch parsing for switches on the command |
292 | line after the script name but before any filename arguments (or before |
293 | a B<-->). Any switch found there is removed from @ARGV and sets the |
294 | corresponding variable in the Perl script. The following script |
295 | prints "true" if and only if the script is invoked with a B<-xyz> switch. |
296 | |
297 | #!/usr/bin/perl -s |
298 | if ($xyz) { print "true\n"; } |
299 | |
300 | =item B<-S> |
301 | |
302 | makes Perl use the PATH environment variable to search for the |
303 | script (unless the name of the script starts with a slash). Typically |
304 | this is used to emulate #! startup on machines that don't support #!, |
305 | in the following manner: |
306 | |
307 | #!/usr/bin/perl |
308 | eval "exec /usr/bin/perl -S $0 $*" |
309 | if $running_under_some_shell; |
310 | |
311 | The system ignores the first line and feeds the script to /bin/sh, |
312 | which proceeds to try to execute the Perl script as a shell script. |
313 | The shell executes the second line as a normal shell command, and thus |
314 | starts up the Perl interpreter. On some systems $0 doesn't always |
315 | contain the full pathname, so the B<-S> tells Perl to search for the |
316 | script if necessary. After Perl locates the script, it parses the |
317 | lines and ignores them because the variable $running_under_some_shell |
318 | is never true. A better construct than C<$*> would be C<${1+"$@"}>, which |
319 | handles embedded spaces and such in the filenames, but doesn't work if |
320 | the script is being interpreted by csh. In order to start up sh rather |
321 | than csh, some systems may have to replace the #! line with a line |
322 | containing just a colon, which will be politely ignored by Perl. Other |
323 | systems can't control that, and need a totally devious construct that |
324 | will work under any of csh, sh or Perl, such as the following: |
325 | |
326 | eval '(exit $?0)' && eval 'exec /usr/bin/perl -S $0 ${1+"$@"}' |
327 | & eval 'exec /usr/bin/perl -S $0 $argv:q' |
328 | if 0; |
329 | |
330 | =item B<-T> |
331 | |
332 | forces "taint" checks to be turned on. Ordinarily these checks are |
333 | done only when running setuid or setgid. See L<perlsec>. |
334 | |
335 | =item B<-u> |
336 | |
337 | causes Perl to dump core after compiling your script. You can then |
338 | take this core dump and turn it into an executable file by using the |
339 | B<undump> program (not supplied). This speeds startup at the expense of |
340 | some disk space (which you can minimize by stripping the executable). |
341 | (Still, a "hello world" executable comes out to about 200K on my |
342 | machine.) If you want to execute a portion of your script before dumping, |
343 | use the dump() operator instead. Note: availability of B<undump> is |
344 | platform specific and may not be available for a specific port of |
345 | Perl. |
346 | |
347 | =item B<-U> |
348 | |
349 | allows Perl to do unsafe operations. Currently the only "unsafe" |
350 | operations are the unlinking of directories while running as superuser, |
351 | and running setuid programs with fatal taint checks turned into |
352 | warnings. |
353 | |
354 | =item B<-v> |
355 | |
356 | prints the version and patchlevel of your Perl executable. |
357 | |
358 | =item B<-w> |
359 | |
360 | prints warnings about identifiers that are mentioned only once, and |
361 | scalar variables that are used before being set. Also warns about |
362 | redefined subroutines, and references to undefined filehandles or |
363 | filehandles opened readonly that you are attempting to write on. Also |
364 | warns you if you use values as a number that doesn't look like numbers, using |
748a9306 |
365 | an array as though it were a scalar, if |
366 | your subroutines recurse more than 100 deep, and innumerable other things. |
a0d0e21e |
367 | See L<perldiag> and L<perltrap>. |
368 | |
369 | =item B<-x> I<directory> |
370 | |
371 | tells Perl that the script is embedded in a message. Leading |
372 | garbage will be discarded until the first line that starts with #! and |
373 | contains the string "perl". Any meaningful switches on that line will |
374 | be applied (but only one group of switches, as with normal #! |
375 | processing). If a directory name is specified, Perl will switch to |
376 | that directory before running the script. The B<-x> switch only |
377 | controls the the disposal of leading garbage. The script must be |
378 | terminated with C<__END__> if there is trailing garbage to be ignored (the |
379 | script can process any or all of the trailing garbage via the DATA |
380 | filehandle if desired). |
381 | |
382 | |
383 | =back |