3 perlfaq5 - Files and Formats
7 This section deals with I/O and the "f" issues: filehandles, flushing,
10 =head2 How do I flush/unbuffer an output filehandle? Why must I do this?
11 X<flush> X<buffer> X<unbuffer> X<autoflush>
13 (contributed by brian d foy)
15 You might like to read Mark Jason Dominus's "Suffering From Buffering"
16 at L<http://perl.plover.com/FAQs/Buffering.html> .
18 Perl normally buffers output so it doesn't make a system call for every
19 bit of output. By saving up output, it makes fewer expensive system calls.
20 For instance, in this little bit of code, you want to print a dot to the
21 screen for every line you process to watch the progress of your program.
22 Instead of seeing a dot for every line, Perl buffers the output and you
23 have a long wait before you see a row of 50 dots all at once:
25 # long wait, then row of dots all at once
28 print "\n" unless ++$count % 50;
30 #... expensive line processing operations
33 To get around this, you have to unbuffer the output filehandle, in this
34 case, C<STDOUT>. You can set the special variable C<$|> to a true value
35 (mnemonic: making your filehandles "piping hot"):
39 # dot shown immediately
42 print "\n" unless ++$count % 50;
44 #... expensive line processing operations
47 The C<$|> is one of the per-filehandle special variables, so each
48 filehandle has its own copy of its value. If you want to merge
49 standard output and standard error for instance, you have to unbuffer
50 each (although STDERR might be unbuffered by default):
53 my $previous_default = select(STDOUT); # save previous default
54 $|++; # autoflush STDOUT
56 $|++; # autoflush STDERR, to be sure
57 select($previous_default); # restore previous default
60 # now should alternate . and +
65 print STDOUT "\n" unless ++$count % 25;
68 Besides the C<$|> special variable, you can use C<binmode> to give
69 your filehandle a C<:unix> layer, which is unbuffered:
71 binmode( STDOUT, ":unix" );
76 print "\n" unless ++$count % 50;
79 For more information on output layers, see the entries for C<binmode>
80 and L<open> in L<perlfunc>, and the L<PerlIO> module documentation.
82 If you are using L<IO::Handle> or one of its subclasses, you can
83 call the C<autoflush> method to change the settings of the
87 open my( $io_fh ), ">", "output.txt";
90 The L<IO::Handle> objects also have a C<flush> method. You can flush
91 the buffer any time you want without auto-buffering
95 =head2 How do I change, delete, or insert a line in a file, or append to the beginning of a file?
98 (contributed by brian d foy)
100 The basic idea of inserting, changing, or deleting a line from a text
101 file involves reading and printing the file to the point you want to
102 make the change, making the change, then reading and printing the rest
103 of the file. Perl doesn't provide random access to lines (especially
104 since the record input separator, C<$/>, is mutable), although modules
105 such as L<Tie::File> can fake it.
107 A Perl program to do these tasks takes the basic form of opening a
108 file, printing its lines, then closing the file:
110 open my $in, '<', $file or die "Can't read old file: $!";
111 open my $out, '>', "$file.new" or die "Can't write new file: $!";
119 Within that basic form, add the parts that you need to insert, change,
122 To prepend lines to the beginning, print those lines before you enter
123 the loop that prints the existing lines.
125 open my $in, '<', $file or die "Can't read old file: $!";
126 open my $out, '>', "$file.new" or die "Can't write new file: $!";
128 print $out "# Add this line to the top\n"; # <--- HERE'S THE MAGIC
136 To change existing lines, insert the code to modify the lines inside
137 the C<while> loop. In this case, the code finds all lowercased
138 versions of "perl" and uppercases them. The happens for every line, so
139 be sure that you're supposed to do that on every line!
141 open my $in, '<', $file or die "Can't read old file: $!";
142 open my $out, '>', "$file.new" or die "Can't write new file: $!";
144 print $out "# Add this line to the top\n";
153 To change only a particular line, the input line number, C<$.>, is
154 useful. First read and print the lines up to the one you want to
155 change. Next, read the single line you want to change, change it, and
156 print it. After that, read the rest of the lines and print those:
158 while( <$in> ) { # print the lines before the change
160 last if $. == 4; # line number before change
164 $line =~ s/\b(perl)\b/Perl/g;
167 while( <$in> ) { # print the rest of the lines
171 To skip lines, use the looping controls. The C<next> in this example
172 skips comment lines, and the C<last> stops all processing once it
173 encounters either C<__END__> or C<__DATA__>.
176 next if /^\s+#/; # skip comment lines
177 last if /^__(END|DATA)__$/; # stop at end of code marker
181 Do the same sort of thing to delete a particular line by using C<next>
182 to skip the lines you don't want to show up in the output. This
183 example skips every fifth line:
190 If, for some odd reason, you really want to see the whole file at once
191 rather than processing line-by-line, you can slurp it in (as long as
192 you can fit the whole thing in memory!):
194 open my $in, '<', $file or die "Can't read old file: $!"
195 open my $out, '>', "$file.new" or die "Can't write new file: $!";
197 my @lines = do { local $/; <$in> }; # slurp!
203 Modules such as L<File::Slurp> and L<Tie::File> can help with that
204 too. If you can, however, avoid reading the entire file at once. Perl
205 won't give that memory back to the operating system until the process
208 You can also use Perl one-liners to modify a file in-place. The
209 following changes all 'Fred' to 'Barney' in F<inFile.txt>, overwriting
210 the file with the new contents. With the C<-p> switch, Perl wraps a
211 C<while> loop around the code you specify with C<-e>, and C<-i> turns
212 on in-place editing. The current line is in C<$_>. With C<-p>, Perl
213 automatically prints the value of C<$_> at the end of the loop. See
214 L<perlrun> for more details.
216 perl -pi -e 's/Fred/Barney/' inFile.txt
218 To make a backup of C<inFile.txt>, give C<-i> a file extension to add:
220 perl -pi.bak -e 's/Fred/Barney/' inFile.txt
222 To change only the fifth line, you can add a test checking C<$.>, the
223 input line number, then only perform the operation when the test
226 perl -pi -e 's/Fred/Barney/ if $. == 5' inFile.txt
228 To add lines before a certain line, you can add a line (or lines!)
229 before Perl prints C<$_>:
231 perl -pi -e 'print "Put before third line\n" if $. == 3' inFile.txt
233 You can even add a line to the beginning of a file, since the current
234 line prints at the end of the loop:
236 perl -pi -e 'print "Put before first line\n" if $. == 1' inFile.txt
238 To insert a line after one already in the file, use the C<-n> switch.
239 It's just like C<-p> except that it doesn't print C<$_> at the end of
240 the loop, so you have to do that yourself. In this case, print C<$_>
241 first, then print the line that you want to add.
243 perl -ni -e 'print; print "Put after fifth line\n" if $. == 5' inFile.txt
245 To delete lines, only print the ones that you want.
247 perl -ni -e 'print unless /d/' inFile.txt
251 perl -pi -e 'next unless /d/' inFile.txt
253 =head2 How do I count the number of lines in a file?
254 X<file, counting lines> X<lines> X<line>
256 (contributed by brian d foy)
258 Conceptually, the easiest way to count the lines in a file is to
259 simply read them and count them:
262 while( <$fh> ) { $count++; }
264 You don't really have to count them yourself, though, since Perl
265 already does that with the C<$.> variable, which is the current line
266 number from the last filehandle read:
271 If you want to use C<$.>, you can reduce it to a simple one-liner,
274 % perl -lne '} print $.; {' file
276 % perl -lne 'END { print $. }' file
278 Those can be rather inefficient though. If they aren't fast enough for
279 you, you might just read chunks of data and count the number of
283 open my($fh), '<:raw', $filename or die "Can't open $filename: $!";
284 while( sysread $fh, $buffer, 4096 ) {
285 $lines += ( $buffer =~ tr/\n// );
289 However, that doesn't work if the line ending isn't a newline. You
290 might change that C<tr///> to a C<s///> so you can count the number of
291 times the input record separator, C<$/>, shows up:
294 open my($fh), '<:raw', $filename or die "Can't open $filename: $!";
295 while( sysread $fh, $buffer, 4096 ) {
296 $lines += ( $buffer =~ s|$/||g; );
300 If you don't mind shelling out, the C<wc> command is usually the
301 fastest, even with the extra interprocess overhead. Ensure that you
302 have an untainted filename though:
309 if( $filename =~ /^([0-9a-z_.]+)\z/ ) {
310 $lines = `/usr/bin/wc -l $1`
314 =head2 How do I delete the last N lines from a file?
317 (contributed by brian d foy)
319 The easiest conceptual solution is to count the lines in the
320 file then start at the beginning and print the number of lines
321 (minus the last N) to a new file.
323 Most often, the real question is how you can delete the last N lines
324 without making more than one pass over the file, or how to do it
325 without a lot of copying. The easy concept is the hard reality when
326 you might have millions of lines in your file.
328 One trick is to use L<File::ReadBackwards>, which starts at the end of
329 the file. That module provides an object that wraps the real filehandle
330 to make it easy for you to move around the file. Once you get to the
331 spot you need, you can get the actual filehandle and work with it as
332 normal. In this case, you get the file position at the end of the last
333 line you want to keep and truncate the file to that point:
335 use File::ReadBackwards;
337 my $filename = 'test.txt';
338 my $Lines_to_truncate = 2;
340 my $bw = File::ReadBackwards->new( $filename )
341 or die "Could not read backwards in [$filename]: $!";
343 my $lines_from_end = 0;
344 until( $bw->eof or $lines_from_end == $Lines_to_truncate ) {
345 print "Got: ", $bw->readline;
349 truncate( $filename, $bw->tell );
351 The L<File::ReadBackwards> module also has the advantage of setting
352 the input record separator to a regular expression.
354 You can also use the L<Tie::File> module which lets you access
355 the lines through a tied array. You can use normal array operations
356 to modify your file, including setting the last index and using
359 =head2 How can I use Perl's C<-i> option from within a program?
362 C<-i> sets the value of Perl's C<$^I> variable, which in turn affects
363 the behavior of C<< <> >>; see L<perlrun> for more details. By
364 modifying the appropriate variables directly, you can get the same
365 behavior within a larger program. For example:
369 local($^I, @ARGV) = ('.orig', glob("*.c"));
372 print "This line should appear at the top of each file\n";
374 s/\b(p)earl\b/${1}erl/i; # Correct typos, preserving case
376 close ARGV if eof; # Reset $.
379 # $^I and @ARGV return to their old values here
381 This block modifies all the C<.c> files in the current directory,
382 leaving a backup of the original data from each file in a new
385 =head2 How can I copy a file?
386 X<copy> X<file, copy> X<File::Copy>
388 (contributed by brian d foy)
390 Use the L<File::Copy> module. It comes with Perl and can do a
391 true copy across file systems, and it does its magic in
396 copy( $original, $new_copy ) or die "Copy failed: $!";
398 If you can't use L<File::Copy>, you'll have to do the work yourself:
399 open the original file, open the destination file, then print
400 to the destination file as you read the original. You also have to
401 remember to copy the permissions, owner, and group to the new file.
403 =head2 How do I make a temporary file name?
406 If you don't need to know the name of the file, you can use C<open()>
407 with C<undef> in place of the file name. In Perl 5.8 or later, the
408 C<open()> function creates an anonymous temporary file:
410 open my $tmp, '+>', undef or die $!;
412 Otherwise, you can use the File::Temp module.
414 use File::Temp qw/ tempfile tempdir /;
416 my $dir = tempdir( CLEANUP => 1 );
417 ($fh, $filename) = tempfile( DIR => $dir );
419 # or if you don't need to know the filename
421 my $fh = tempfile( DIR => $dir );
423 The File::Temp has been a standard module since Perl 5.6.1. If you
424 don't have a modern enough Perl installed, use the C<new_tmpfile>
425 class method from the IO::File module to get a filehandle opened for
426 reading and writing. Use it if you don't need to know the file's name:
429 my $fh = IO::File->new_tmpfile()
430 or die "Unable to make new temporary file: $!";
432 If you're committed to creating a temporary file by hand, use the
433 process ID and/or the current time-value. If you need to have many
434 temporary files in one process, use a counter:
438 my $temp_dir = -d '/tmp' ? '/tmp' : $ENV{TMPDIR} || $ENV{TEMP};
439 my $base_name = sprintf "%s/%d-%d-0000", $temp_dir, $$, time;
444 until( defined(fileno($fh)) || $count++ > 100 ) {
445 $base_name =~ s/-(\d+)$/"-" . (1 + $1)/e;
446 # O_EXCL is required for security reasons.
447 sysopen $fh, $base_name, O_WRONLY|O_EXCL|O_CREAT;
450 if( defined fileno($fh) ) {
451 return ($fh, $base_name);
459 =head2 How can I manipulate fixed-record-length files?
460 X<fixed-length> X<file, fixed-length records>
462 The most efficient way is using L<pack()|perlfunc/"pack"> and
463 L<unpack()|perlfunc/"unpack">. This is faster than using
464 L<substr()|perlfunc/"substr"> when taking many, many strings. It is
465 slower for just a few.
467 Here is a sample chunk of code to break up and put back together again
468 some fixed-format input lines, in this case from the output of a normal,
472 # 15158 p5 T 0:00 perl /home/tchrist/scripts/now-what
473 my $PS_T = 'A6 A4 A7 A5 A*';
474 open my $ps, '-|', 'ps';
476 my @fields = qw( pid tt stat time command );
479 @process{@fields} = unpack($PS_T, $_);
480 for my $field ( @fields ) {
481 print "$field: <$process{$field}>\n";
483 print 'line=', pack($PS_T, @process{@fields} ), "\n";
486 We've used a hash slice in order to easily handle the fields of each row.
487 Storing the keys in an array makes it easy to operate on them as a
488 group or loop over them with C<for>. It also avoids polluting the program
489 with global variables and using symbolic references.
491 =head2 How can I make a filehandle local to a subroutine? How do I pass filehandles between subroutines? How do I make an array of filehandles?
492 X<filehandle, local> X<filehandle, passing> X<filehandle, reference>
494 As of perl5.6, open() autovivifies file and directory handles
495 as references if you pass it an uninitialized scalar variable.
496 You can then pass these references just like any other scalar,
497 and use them in the place of named handles.
499 open my $fh, $file_name;
501 open local $fh, $file_name;
503 print $fh "Hello World!\n";
507 If you like, you can store these filehandles in an array or a hash.
508 If you access them directly, they aren't simple scalars and you
509 need to give C<print> a little help by placing the filehandle
510 reference in braces. Perl can only figure it out on its own when
511 the filehandle reference is a simple scalar.
513 my @fhs = ( $fh1, $fh2, $fh3 );
515 for( $i = 0; $i <= $#fhs; $i++ ) {
516 print {$fhs[$i]} "just another Perl answer, \n";
519 Before perl5.6, you had to deal with various typeglob idioms
520 which you may see in older code.
522 open FILE, "> $filename";
523 process_typeglob( *FILE );
524 process_reference( \*FILE );
526 sub process_typeglob { local *FH = shift; print FH "Typeglob!" }
527 sub process_reference { local $fh = shift; print $fh "Reference!" }
529 If you want to create many anonymous handles, you should
530 check out the Symbol or IO::Handle modules.
532 =head2 How can I use a filehandle indirectly?
533 X<filehandle, indirect>
535 An indirect filehandle is the use of something other than a symbol
536 in a place that a filehandle is expected. Here are ways
537 to get indirect filehandles:
539 $fh = SOME_FH; # bareword is strict-subs hostile
540 $fh = "SOME_FH"; # strict-refs hostile; same package only
541 $fh = *SOME_FH; # typeglob
542 $fh = \*SOME_FH; # ref to typeglob (bless-able)
543 $fh = *SOME_FH{IO}; # blessed IO::Handle from *SOME_FH typeglob
545 Or, you can use the C<new> method from one of the IO::* modules to
546 create an anonymous filehandle and store that in a scalar variable.
548 use IO::Handle; # 5.004 or higher
549 my $fh = IO::Handle->new();
551 Then use any of those as you would a normal filehandle. Anywhere that
552 Perl is expecting a filehandle, an indirect filehandle may be used
553 instead. An indirect filehandle is just a scalar variable that contains
554 a filehandle. Functions like C<print>, C<open>, C<seek>, or
555 the C<< <FH> >> diamond operator will accept either a named filehandle
556 or a scalar variable containing one:
558 ($ifh, $ofh, $efh) = (*STDIN, *STDOUT, *STDERR);
559 print $ofh "Type it: ";
561 print $efh "What was that: $got";
563 If you're passing a filehandle to a function, you can write
564 the function in two ways:
568 print $fh "Sending to indirect filehandle\n";
571 Or it can localize a typeglob and use the filehandle directly:
575 print FH "Sending to localized filehandle\n";
578 Both styles work with either objects or typeglobs of real filehandles.
579 (They might also work with strings under some circumstances, but this
585 In the examples above, we assigned the filehandle to a scalar variable
586 before using it. That is because only simple scalar variables, not
587 expressions or subscripts of hashes or arrays, can be used with
588 built-ins like C<print>, C<printf>, or the diamond operator. Using
589 something other than a simple scalar variable as a filehandle is
590 illegal and won't even compile:
592 my @fd = (*STDIN, *STDOUT, *STDERR);
593 print $fd[1] "Type it: "; # WRONG
594 my $got = <$fd[0]> # WRONG
595 print $fd[2] "What was that: $got"; # WRONG
597 With C<print> and C<printf>, you get around this by using a block and
598 an expression where you would place the filehandle:
600 print { $fd[1] } "funny stuff\n";
601 printf { $fd[1] } "Pity the poor %x.\n", 3_735_928_559;
602 # Pity the poor deadbeef.
604 That block is a proper block like any other, so you can put more
605 complicated code there. This sends the message out to one of two places:
607 my $ok = -x "/bin/cat";
608 print { $ok ? $fd[1] : $fd[2] } "cat stat $ok\n";
609 print { $fd[ 1+ ($ok || 0) ] } "cat stat $ok\n";
611 This approach of treating C<print> and C<printf> like object methods
612 calls doesn't work for the diamond operator. That's because it's a
613 real operator, not just a function with a comma-less argument. Assuming
614 you've been storing typeglobs in your structure as we did above, you
615 can use the built-in function named C<readline> to read a record just
616 as C<< <> >> does. Given the initialization shown above for @fd, this
617 would work, but only because readline() requires a typeglob. It doesn't
618 work with objects or strings, which might be a bug we haven't fixed yet.
620 $got = readline($fd[0]);
622 Let it be noted that the flakiness of indirect filehandles is not
623 related to whether they're strings, typeglobs, objects, or anything else.
624 It's the syntax of the fundamental operators. Playing the object
625 game doesn't help you at all here.
627 =head2 How can I set up a footer format to be used with write()?
630 There's no builtin way to do this, but L<perlform> has a couple of
631 techniques to make it possible for the intrepid hacker.
633 =head2 How can I write() into a string?
634 X<write, into a string>
636 (contributed by brian d foy)
638 If you want to C<write> into a string, you just have to <open> a
639 filehandle to a string, which Perl has been able to do since Perl 5.6:
641 open FH, '>', \my $string;
644 Since you want to be a good programmer, you probably want to use a lexical
645 filehandle, even though formats are designed to work with bareword filehandles
646 since the default format names take the filehandle name. However, you can
647 control this with some Perl special per-filehandle variables: C<$^>, which
648 names the top-of-page format, and C<$~> which shows the line format. You have
649 to change the default filehandle to set these variables:
651 open my($fh), '>', \my $string;
653 { # set per-filehandle variables
654 my $old_fh = select( $fh );
665 @## @<<< @<<<<<<<<<<<<<<
669 Although write can work with lexical or package variables, whatever variables
670 you use have to scope in the format. That most likely means you'll want to
671 localize some package variables:
674 local( $id, $type, $name ) = qw( 12 cat Buster );
680 There are also some tricks that you can play with C<formline> and the
681 accumulator variable C<$^A>, but you lose a lot of the value of formats
682 since C<formline> won't handle paging and so on. You end up reimplementing
683 formats when you use them.
685 =head2 How can I open a filehandle to a string?
686 X<string> X<open> X<IO::String> X<filehandle>
688 (contributed by Peter J. Holzer, hjp-usenet2@hjp.at)
690 Since Perl 5.8.0 a file handle referring to a string can be created by
691 calling open with a reference to that string instead of the filename.
692 This file handle can then be used to read from or write to the string:
694 open(my $fh, '>', \$string) or die "Could not open string for writing";
696 print $fh "bar\n"; # $string now contains "foo\nbar\n"
698 open(my $fh, '<', \$string) or die "Could not open string for reading";
699 my $x = <$fh>; # $x now contains "foo\n"
701 With older versions of Perl, the L<IO::String> module provides similar
704 =head2 How can I output my numbers with commas added?
707 (contributed by brian d foy and Benjamin Goldberg)
709 You can use L<Number::Format> to separate places in a number.
710 It handles locale information for those of you who want to insert
711 full stops instead (or anything else that they want to use,
714 This subroutine will add commas to your number:
718 1 while s/^([-+]?\d+)(\d{3})/$1,$2/;
722 This regex from Benjamin Goldberg will add commas to numbers:
724 s/(^[-+]?\d+?(?=(?>(?:\d{3})+)(?!\d))|\G\d{3}(?=\d))/$1,/g;
726 It is easier to see with comments:
729 ^[-+]? # beginning of number.
730 \d+? # first digits before first comma
731 (?= # followed by, (but not included in the match) :
732 (?>(?:\d{3})+) # some positive multiple of three digits.
733 (?!\d) # an *exact* multiple, not x * 3 + 1 or whatever.
736 \G\d{3} # after the last group, get three digits
737 (?=\d) # but they have to have more digits after them.
740 =head2 How can I translate tildes (~) in a filename?
741 X<tilde> X<tilde expansion>
743 Use the E<lt>E<gt> (C<glob()>) operator, documented in L<perlfunc>.
744 Versions of Perl older than 5.6 require that you have a shell
745 installed that groks tildes. Later versions of Perl have this feature
746 built in. The L<File::KGlob> module (available from CPAN) gives more
747 portable glob functionality.
749 Within Perl, you may use this directly:
752 ^ ~ # find a leading tilde
754 [^/] # a non-slash character
755 * # repeated 0 or more times (0 means me)
760 : ( $ENV{HOME} || $ENV{LOGDIR} )
763 =head2 How come when I open a file read-write it wipes it out?
764 X<clobber> X<read-write> X<clobbering> X<truncate> X<truncating>
766 Because you're using something like this, which truncates the file
767 I<then> gives you read-write access:
769 open my $fh, '+>', '/path/name'; # WRONG (almost always)
771 Whoops. You should instead use this, which will fail if the file
774 open my $fh, '+<', '/path/name'; # open for update
776 Using ">" always clobbers or creates. Using "<" never does
777 either. The "+" doesn't change this.
779 Here are examples of many kinds of file opens. Those using C<sysopen>
780 all assume that you've pulled in the constants from L<Fcntl>:
784 To open file for reading:
786 open my $fh, '<', $path or die $!;
787 sysopen my $fh, $path, O_RDONLY or die $!;
789 To open file for writing, create new file if needed or else truncate old file:
791 open my $fh, '>', $path or die $!;
792 sysopen my $fh, $path, O_WRONLY|O_TRUNC|O_CREAT or die $!;
793 sysopen my $fh, $path, O_WRONLY|O_TRUNC|O_CREAT, 0666 or die $!;
795 To open file for writing, create new file, file must not exist:
797 sysopen my $fh, $path, O_WRONLY|O_EXCL|O_CREAT or die $!;
798 sysopen my $fh, $path, O_WRONLY|O_EXCL|O_CREAT, 0666 or die $!;
800 To open file for appending, create if necessary:
802 open my $fh, '>>' $path or die $!;
803 sysopen my $fh, $path, O_WRONLY|O_APPEND|O_CREAT or die $!;
804 sysopen my $fh, $path, O_WRONLY|O_APPEND|O_CREAT, 0666 or die $!;
806 To open file for appending, file must exist:
808 sysopen my $fh, $path, O_WRONLY|O_APPEND or die $!;
810 To open file for update, file must exist:
812 open my $fh, '+<', $path or die $!;
813 sysopen my $fh, $path, O_RDWR or die $!;
815 To open file for update, create file if necessary:
817 sysopen my $fh, $path, O_RDWR|O_CREAT or die $!;
818 sysopen my $fh, $path, O_RDWR|O_CREAT, 0666 or die $!;
820 To open file for update, file must not exist:
822 sysopen my $fh, $path, O_RDWR|O_EXCL|O_CREAT or die $!;
823 sysopen my $fh, $path, O_RDWR|O_EXCL|O_CREAT, 0666 or die $!;
825 To open a file without blocking, creating if necessary:
827 sysopen my $fh, '/foo/somefile', O_WRONLY|O_NDELAY|O_CREAT
828 or die "can't open /foo/somefile: $!":
830 Be warned that neither creation nor deletion of files is guaranteed to
831 be an atomic operation over NFS. That is, two processes might both
832 successfully create or unlink the same file! Therefore O_EXCL
833 isn't as exclusive as you might wish.
835 See also L<perlopentut>.
837 =head2 Why do I sometimes get an "Argument list too long" when I use E<lt>*E<gt>?
838 X<argument list too long>
840 The C<< <> >> operator performs a globbing operation (see above).
841 In Perl versions earlier than v5.6.0, the internal glob() operator forks
842 csh(1) to do the actual glob expansion, but
843 csh can't handle more than 127 items and so gives the error message
844 C<Argument list too long>. People who installed tcsh as csh won't
845 have this problem, but their users may be surprised by it.
847 To get around this, either upgrade to Perl v5.6.0 or later, do the glob
848 yourself with readdir() and patterns, or use a module like L<File::Glob>,
849 one that doesn't use the shell to do globbing.
851 =head2 Is there a leak/bug in glob()?
854 (contributed by brian d foy)
856 Starting with Perl 5.6.0, C<glob> is implemented internally rather
857 than relying on an external resource. As such, memory issues with
858 C<glob> aren't a problem in modern perls.
860 =head2 How can I open a file with a leading ">" or trailing blanks?
861 X<filename, special characters>
863 (contributed by Brian McCauley)
865 The special two-argument form of Perl's open() function ignores
866 trailing blanks in filenames and infers the mode from certain leading
867 characters (or a trailing "|"). In older versions of Perl this was the
868 only version of open() and so it is prevalent in old code and books.
870 Unless you have a particular reason to use the two-argument form you
871 should use the three-argument form of open() which does not treat any
872 characters in the filename as special.
874 open my $fh, "<", " file "; # filename is " file "
875 open my $fh, ">", ">file"; # filename is ">file"
877 =head2 How can I reliably rename a file?
878 X<rename> X<mv> X<move> X<file, rename>
880 If your operating system supports a proper mv(1) utility or its
881 functional equivalent, this works:
883 rename($old, $new) or system("mv", $old, $new);
885 It may be more portable to use the L<File::Copy> module instead.
886 You just copy to the new file to the new name (checking return
887 values), then delete the old one. This isn't really the same
888 semantically as a C<rename()>, which preserves meta-information like
889 permissions, timestamps, inode info, etc.
891 =head2 How can I lock a file?
892 X<lock> X<file, lock> X<flock>
894 Perl's builtin flock() function (see L<perlfunc> for details) will call
895 flock(2) if that exists, fcntl(2) if it doesn't (on perl version 5.004 and
896 later), and lockf(3) if neither of the two previous system calls exists.
897 On some systems, it may even use a different form of native locking.
898 Here are some gotchas with Perl's flock():
904 Produces a fatal error if none of the three system calls (or their
905 close equivalent) exists.
909 lockf(3) does not provide shared locking, and requires that the
910 filehandle be open for writing (or appending, or read/writing).
914 Some versions of flock() can't lock files over a network (e.g. on NFS file
915 systems), so you'd need to force the use of fcntl(2) when you build Perl.
916 But even this is dubious at best. See the flock entry of L<perlfunc>
917 and the F<INSTALL> file in the source distribution for information on
918 building Perl to do this.
920 Two potentially non-obvious but traditional flock semantics are that
921 it waits indefinitely until the lock is granted, and that its locks are
922 I<merely advisory>. Such discretionary locks are more flexible, but
923 offer fewer guarantees. This means that files locked with flock() may
924 be modified by programs that do not also use flock(). Cars that stop
925 for red lights get on well with each other, but not with cars that don't
926 stop for red lights. See the perlport manpage, your port's specific
927 documentation, or your system-specific local manpages for details. It's
928 best to assume traditional behavior if you're writing portable programs.
929 (If you're not, you should as always feel perfectly free to write
930 for your own system's idiosyncrasies (sometimes called "features").
931 Slavish adherence to portability concerns shouldn't get in the way of
932 your getting your job done.)
934 For more information on file locking, see also
935 L<perlopentut/"File Locking"> if you have it (new for 5.6).
939 =head2 Why can't I just open(FH, "E<gt>file.lock")?
940 X<lock, lockfile race condition>
942 A common bit of code B<NOT TO USE> is this:
944 sleep(3) while -e 'file.lock'; # PLEASE DO NOT USE
945 open my $lock, '>', 'file.lock'; # THIS BROKEN CODE
947 This is a classic race condition: you take two steps to do something
948 which must be done in one. That's why computer hardware provides an
949 atomic test-and-set instruction. In theory, this "ought" to work:
951 sysopen my $fh, "file.lock", O_WRONLY|O_EXCL|O_CREAT
952 or die "can't open file.lock: $!";
954 except that lamentably, file creation (and deletion) is not atomic
955 over NFS, so this won't work (at least, not every time) over the net.
956 Various schemes involving link() have been suggested, but
957 these tend to involve busy-wait, which is also less than desirable.
959 =head2 I still don't get locking. I just want to increment the number in the file. How can I do this?
960 X<counter> X<file, counter>
962 Didn't anyone ever tell you web-page hit counters were useless?
963 They don't count number of hits, they're a waste of time, and they serve
964 only to stroke the writer's vanity. It's better to pick a random number;
965 they're more realistic.
967 Anyway, this is what you can do if you can't help yourself.
969 use Fcntl qw(:DEFAULT :flock);
970 sysopen my $fh, "numfile", O_RDWR|O_CREAT or die "can't open numfile: $!";
971 flock $fh, LOCK_EX or die "can't flock numfile: $!";
972 my $num = <$fh> || 0;
973 seek $fh, 0, 0 or die "can't rewind numfile: $!";
974 truncate $fh, 0 or die "can't truncate numfile: $!";
975 (print $fh $num+1, "\n") or die "can't write numfile: $!";
976 close $fh or die "can't close numfile: $!";
978 Here's a much better web-page hit counter:
980 $hits = int( (time() - 850_000_000) / rand(1_000) );
982 If the count doesn't impress your friends, then the code might. :-)
984 =head2 All I want to do is append a small amount of text to the end of a file. Do I still have to use locking?
985 X<append> X<file, append>
987 If you are on a system that correctly implements C<flock> and you use
988 the example appending code from "perldoc -f flock" everything will be
989 OK even if the OS you are on doesn't implement append mode correctly
990 (if such a system exists). So if you are happy to restrict yourself to
991 OSs that implement C<flock> (and that's not really much of a
992 restriction) then that is what you should do.
994 If you know you are only going to use a system that does correctly
995 implement appending (i.e. not Win32) then you can omit the C<seek>
996 from the code in the previous answer.
998 If you know you are only writing code to run on an OS and filesystem
999 that does implement append mode correctly (a local filesystem on a
1000 modern Unix for example), and you keep the file in block-buffered mode
1001 and you write less than one buffer-full of output between each manual
1002 flushing of the buffer then each bufferload is almost guaranteed to be
1003 written to the end of the file in one chunk without getting
1004 intermingled with anyone else's output. You can also use the
1005 C<syswrite> function which is simply a wrapper around your system's
1006 C<write(2)> system call.
1008 There is still a small theoretical chance that a signal will interrupt
1009 the system-level C<write()> operation before completion. There is also
1010 a possibility that some STDIO implementations may call multiple system
1011 level C<write()>s even if the buffer was empty to start. There may be
1012 some systems where this probability is reduced to zero, and this is
1013 not a concern when using C<:perlio> instead of your system's STDIO.
1015 =head2 How do I randomly update a binary file?
1016 X<file, binary patch>
1018 If you're just trying to patch a binary, in many cases something as
1019 simple as this works:
1021 perl -i -pe 's{window manager}{window mangler}g' /usr/bin/emacs
1023 However, if you have fixed sized records, then you might do something more
1026 my $RECSIZE = 220; # size of record, in bytes
1027 my $recno = 37; # which record to update
1028 open my $fh, '+<', 'somewhere' or die "can't update somewhere: $!";
1029 seek $fh, $recno * $RECSIZE, 0;
1030 read $fh, $record, $RECSIZE == $RECSIZE or die "can't read record $recno: $!";
1032 seek $fh, -$RECSIZE, 1;
1036 Locking and error checking are left as an exercise for the reader.
1037 Don't forget them or you'll be quite sorry.
1039 =head2 How do I get a file's timestamp in perl?
1040 X<timestamp> X<file, timestamp>
1042 If you want to retrieve the time at which the file was last read,
1043 written, or had its meta-data (owner, etc) changed, you use the B<-A>,
1044 B<-M>, or B<-C> file test operations as documented in L<perlfunc>.
1045 These retrieve the age of the file (measured against the start-time of
1046 your program) in days as a floating point number. Some platforms may
1047 not have all of these times. See L<perlport> for details. To retrieve
1048 the "raw" time in seconds since the epoch, you would call the stat
1049 function, then use C<localtime()>, C<gmtime()>, or
1050 C<POSIX::strftime()> to convert this into human-readable form.
1054 my $write_secs = (stat($file))[9];
1055 printf "file %s updated at %s\n", $file,
1056 scalar localtime($write_secs);
1058 If you prefer something more legible, use the File::stat module
1059 (part of the standard distribution in version 5.004 and later):
1061 # error checking left as an exercise for reader.
1063 use Time::localtime;
1064 my $date_string = ctime(stat($file)->mtime);
1065 print "file $file updated at $date_string\n";
1067 The POSIX::strftime() approach has the benefit of being,
1068 in theory, independent of the current locale. See L<perllocale>
1071 =head2 How do I set a file's timestamp in perl?
1072 X<timestamp> X<file, timestamp>
1074 You use the utime() function documented in L<perlfunc/utime>.
1075 By way of example, here's a little program that copies the
1076 read and write times from its first argument to all the rest
1080 die "usage: cptimes timestamp_file other_files ...\n";
1082 my $timestamp = shift;
1083 my($atime, $mtime) = (stat($timestamp))[8,9];
1084 utime $atime, $mtime, @ARGV;
1086 Error checking is, as usual, left as an exercise for the reader.
1088 The perldoc for utime also has an example that has the same
1089 effect as touch(1) on files that I<already exist>.
1091 Certain file systems have a limited ability to store the times
1092 on a file at the expected level of precision. For example, the
1093 FAT and HPFS filesystem are unable to create dates on files with
1094 a finer granularity than two seconds. This is a limitation of
1095 the filesystems, not of utime().
1097 =head2 How do I print to more than one file at once?
1098 X<print, to multiple files>
1100 To connect one filehandle to several output filehandles,
1101 you can use the L<IO::Tee> or L<Tie::FileHandle::Multiplex> modules.
1103 If you only have to do this once, you can print individually
1106 for my $fh ($fh1, $fh2, $fh3) { print $fh "whatever\n" }
1108 =head2 How can I read in an entire file all at once?
1109 X<slurp> X<file, slurping>
1111 The customary Perl approach for processing all the lines in a file is to
1112 do so one line at a time:
1114 open my $input, '<', $file or die "can't open $file: $!";
1117 # do something with $_
1119 close $input or die "can't close $file: $!";
1121 This is tremendously more efficient than reading the entire file into
1122 memory as an array of lines and then processing it one element at a time,
1123 which is often--if not almost always--the wrong approach. Whenever
1124 you see someone do this:
1126 my @lines = <INPUT>;
1128 You should think long and hard about why you need everything loaded at
1129 once. It's just not a scalable solution.
1131 If you "mmap" the file with the File::Map module from
1132 CPAN, you can virtually load the entire file into a
1133 string without actually storing it in memory:
1135 use File::Map qw(map_file);
1137 map_file my $string, $filename;
1139 Once mapped, you can treat C<$string> as you would any other string.
1140 Since you don't necessarily have to load the data, mmap-ing can be
1141 very fast and may not increase your memory footprint.
1143 You might also find it more
1144 fun to use the standard L<Tie::File> module, or the L<DB_File> module's
1145 C<$DB_RECNO> bindings, which allow you to tie an array to a file so that
1146 accessing an element of the array actually accesses the corresponding
1149 If you want to load the entire file, you can use the L<File::Slurp>
1150 module to do it in one one simple and efficient step:
1154 my $all_of_it = read_file($filename); # entire file in scalar
1155 my @all_lines = read_file($filename); # one line per element
1157 Or you can read the entire file contents into a scalar like this:
1162 open my $fh, '<', $file or die "can't open $file: $!";
1166 That temporarily undefs your record separator, and will automatically
1167 close the file at block exit. If the file is already open, just use this:
1169 my $var = do { local $/; <$fh> };
1171 You can also use a localized C<@ARGV> to eliminate the C<open>:
1173 my $var = do { local( @ARGV, $/ ) = $file; <> };
1175 For ordinary files you can also use the C<read> function.
1177 read( $fh, $var, -s $fh );
1179 That third argument tests the byte size of the data on the C<$fh> filehandle
1180 and reads that many bytes into the buffer C<$var>.
1182 =head2 How can I read in a file by paragraphs?
1183 X<file, reading by paragraphs>
1185 Use the C<$/> variable (see L<perlvar> for details). You can either
1186 set it to C<""> to eliminate empty paragraphs (C<"abc\n\n\n\ndef">,
1187 for instance, gets treated as two paragraphs and not three), or
1188 C<"\n\n"> to accept empty paragraphs.
1190 Note that a blank line must have no blanks in it. Thus
1191 S<C<"fred\n \nstuff\n\n">> is one paragraph, but C<"fred\n\nstuff\n\n"> is two.
1193 =head2 How can I read a single character from a file? From the keyboard?
1194 X<getc> X<file, reading one character at a time>
1196 You can use the builtin C<getc()> function for most filehandles, but
1197 it won't (easily) work on a terminal device. For STDIN, either use
1198 the Term::ReadKey module from CPAN or use the sample code in
1201 If your system supports the portable operating system programming
1202 interface (POSIX), you can use the following code, which you'll note
1203 turns off echo processing as well.
1216 use POSIX qw(:termios_h);
1218 my ($term, $oterm, $echo, $noecho, $fd_stdin);
1220 my $fd_stdin = fileno(STDIN);
1222 $term = POSIX::Termios->new();
1223 $term->getattr($fd_stdin);
1224 $oterm = $term->getlflag();
1226 $echo = ECHO | ECHOK | ICANON;
1227 $noecho = $oterm & ~$echo;
1230 $term->setlflag($noecho);
1231 $term->setcc(VTIME, 1);
1232 $term->setattr($fd_stdin, TCSANOW);
1236 $term->setlflag($oterm);
1237 $term->setcc(VTIME, 0);
1238 $term->setattr($fd_stdin, TCSANOW);
1244 sysread(STDIN, $key, 1);
1252 The Term::ReadKey module from CPAN may be easier to use. Recent versions
1253 include also support for non-portable systems as well.
1256 open my $tty, '<', '/dev/tty';
1257 print "Gimme a char: ";
1259 my $key = ReadKey 0, $tty;
1261 printf "\nYou said %s, char number %03d\n",
1264 =head2 How can I tell whether there's a character waiting on a filehandle?
1266 The very first thing you should do is look into getting the Term::ReadKey
1267 extension from CPAN. As we mentioned earlier, it now even has limited
1268 support for non-portable (read: not open systems, closed, proprietary,
1269 not POSIX, not Unix, etc.) systems.
1271 You should also check out the Frequently Asked Questions list in
1272 comp.unix.* for things like this: the answer is essentially the same.
1273 It's very system-dependent. Here's one solution that works on BSD
1278 vec($rin, fileno(STDIN), 1) = 1;
1279 return $nfd = select($rin,undef,undef,0);
1282 If you want to find out how many characters are waiting, there's
1283 also the FIONREAD ioctl call to be looked at. The I<h2ph> tool that
1284 comes with Perl tries to convert C include files to Perl code, which
1285 can be C<require>d. FIONREAD ends up defined as a function in the
1286 I<sys/ioctl.ph> file:
1288 require 'sys/ioctl.ph';
1290 $size = pack("L", 0);
1291 ioctl(FH, FIONREAD(), $size) or die "Couldn't call ioctl: $!\n";
1292 $size = unpack("L", $size);
1294 If I<h2ph> wasn't installed or doesn't work for you, you can
1295 I<grep> the include files by hand:
1297 % grep FIONREAD /usr/include/*/*
1298 /usr/include/asm/ioctls.h:#define FIONREAD 0x541B
1300 Or write a small C program using the editor of champions:
1303 #include <sys/ioctl.h>
1305 printf("%#08x\n", FIONREAD);
1308 % cc -o fionread fionread.c
1312 And then hard-code it, leaving porting as an exercise to your successor.
1314 $FIONREAD = 0x4004667f; # XXX: opsys dependent
1316 $size = pack("L", 0);
1317 ioctl(FH, $FIONREAD, $size) or die "Couldn't call ioctl: $!\n";
1318 $size = unpack("L", $size);
1320 FIONREAD requires a filehandle connected to a stream, meaning that sockets,
1321 pipes, and tty devices work, but I<not> files.
1323 =head2 How do I do a C<tail -f> in perl?
1324 X<tail> X<IO::Handle> X<File::Tail> X<clearerr>
1330 The statement C<seek($gw_fh, 0, 1)> doesn't change the current position,
1331 but it does clear the end-of-file condition on the handle, so that the
1332 next C<< <$gw_fh> >> makes Perl try again to read something.
1334 If that doesn't work (it relies on features of your stdio implementation),
1335 then you need something more like this:
1338 for ($curpos = tell($gw_fh); <$gw_fh>; $curpos =tell($gw_fh)) {
1339 # search for some stuff and put it into files
1342 seek($gw_fh, $curpos, 0); # seek to where we had been
1345 If this still doesn't work, look into the C<clearerr> method
1346 from L<IO::Handle>, which resets the error and end-of-file states
1349 There's also a L<File::Tail> module from CPAN.
1351 =head2 How do I dup() a filehandle in Perl?
1354 If you check L<perlfunc/open>, you'll see that several of the ways
1355 to call open() should do the trick. For example:
1357 open my $log, '>>', '/foo/logfile';
1358 open STDERR, '>&', $log;
1360 Or even with a literal numeric descriptor:
1362 my $fd = $ENV{MHCONTEXTFD};
1363 open $mhcontext, "<&=$fd"; # like fdopen(3S)
1365 Note that "<&STDIN" makes a copy, but "<&=STDIN" makes
1366 an alias. That means if you close an aliased handle, all
1367 aliases become inaccessible. This is not true with
1370 Error checking, as always, has been left as an exercise for the reader.
1372 =head2 How do I close a file descriptor by number?
1373 X<file, closing file descriptors> X<POSIX> X<close>
1375 If, for some reason, you have a file descriptor instead of a
1376 filehandle (perhaps you used C<POSIX::open>), you can use the
1377 C<close()> function from the L<POSIX> module:
1381 POSIX::close( $fd );
1383 This should rarely be necessary, as the Perl C<close()> function is to be
1384 used for things that Perl opened itself, even if it was a dup of a
1385 numeric descriptor as with C<MHCONTEXT> above. But if you really have
1386 to, you may be able to do this:
1388 require 'sys/syscall.ph';
1389 my $rc = syscall(&SYS_close, $fd + 0); # must force numeric
1390 die "can't sysclose $fd: $!" unless $rc == -1;
1392 Or, just use the fdopen(3S) feature of C<open()>:
1395 open my $fh, "<&=$fd" or die "Cannot reopen fd=$fd: $!";
1399 =head2 Why can't I use "C:\temp\foo" in DOS paths? Why doesn't `C:\temp\foo.exe` work?
1400 X<filename, DOS issues>
1402 Whoops! You just put a tab and a formfeed into that filename!
1403 Remember that within double quoted strings ("like\this"), the
1404 backslash is an escape character. The full list of these is in
1405 L<perlop/Quote and Quote-like Operators>. Unsurprisingly, you don't
1406 have a file called "c:(tab)emp(formfeed)oo" or
1407 "c:(tab)emp(formfeed)oo.exe" on your legacy DOS filesystem.
1409 Either single-quote your strings, or (preferably) use forward slashes.
1410 Since all DOS and Windows versions since something like MS-DOS 2.0 or so
1411 have treated C</> and C<\> the same in a path, you might as well use the
1412 one that doesn't clash with Perl--or the POSIX shell, ANSI C and C++,
1413 awk, Tcl, Java, or Python, just to mention a few. POSIX paths
1414 are more portable, too.
1416 =head2 Why doesn't glob("*.*") get all the files?
1419 Because even on non-Unix ports, Perl's glob function follows standard
1420 Unix globbing semantics. You'll need C<glob("*")> to get all (non-hidden)
1421 files. This makes glob() portable even to legacy systems. Your
1422 port may include proprietary globbing functions as well. Check its
1423 documentation for details.
1425 =head2 Why does Perl let me delete read-only files? Why does C<-i> clobber protected files? Isn't this a bug in Perl?
1427 This is elaborately and painstakingly described in the
1428 F<file-dir-perms> article in the "Far More Than You Ever Wanted To
1429 Know" collection in L<http://www.cpan.org/misc/olddoc/FMTEYEWTK.tgz> .
1431 The executive summary: learn how your filesystem works. The
1432 permissions on a file say what can happen to the data in that file.
1433 The permissions on a directory say what can happen to the list of
1434 files in that directory. If you delete a file, you're removing its
1435 name from the directory (so the operation depends on the permissions
1436 of the directory, not of the file). If you try to write to the file,
1437 the permissions of the file govern whether you're allowed to.
1439 =head2 How do I select a random line from a file?
1440 X<file, selecting a random line>
1442 Short of loading the file into a database or pre-indexing the lines in
1443 the file, there are a couple of things that you can do.
1445 Here's a reservoir-sampling algorithm from the Camel Book:
1448 rand($.) < 1 && ($line = $_) while <>;
1450 This has a significant advantage in space over reading the whole file
1451 in. You can find a proof of this method in I<The Art of Computer
1452 Programming>, Volume 2, Section 3.4.2, by Donald E. Knuth.
1454 You can use the L<File::Random> module which provides a function
1457 use File::Random qw/random_line/;
1458 my $line = random_line($filename);
1460 Another way is to use the L<Tie::File> module, which treats the entire
1461 file as an array. Simply access a random array element.
1463 =head2 Why do I get weird spaces when I print an array of lines?
1465 (contributed by brian d foy)
1467 If you are seeing spaces between the elements of your array when
1468 you print the array, you are probably interpolating the array in
1471 my @animals = qw(camel llama alpaca vicuna);
1472 print "animals are: @animals\n";
1474 It's the double quotes, not the C<print>, doing this. Whenever you
1475 interpolate an array in a double quote context, Perl joins the
1476 elements with spaces (or whatever is in C<$">, which is a space by
1479 animals are: camel llama alpaca vicuna
1481 This is different than printing the array without the interpolation:
1483 my @animals = qw(camel llama alpaca vicuna);
1484 print "animals are: ", @animals, "\n";
1486 Now the output doesn't have the spaces between the elements because
1487 the elements of C<@animals> simply become part of the list to
1490 animals are: camelllamaalpacavicuna
1492 You might notice this when each of the elements of C<@array> end with
1493 a newline. You expect to print one element per line, but notice that
1494 every line after the first is indented:
1497 this is another line
1498 this is the third line
1500 That extra space comes from the interpolation of the array. If you
1501 don't want to put anything between your array elements, don't use the
1502 array in double quotes. You can send it to print without them:
1506 =head2 How do I traverse a directory tree?
1508 (contributed by brian d foy)
1510 The L<File::Find> module, which comes with Perl, does all of the hard
1511 work to traverse a directory structure. It comes with Perl. You simply
1512 call the C<find> subroutine with a callback subroutine and the
1513 directories you want to traverse:
1517 find( \&wanted, @directories );
1520 # full path in $File::Find::name
1521 # just filename in $_
1522 ... do whatever you want to do ...
1525 The L<File::Find::Closures>, which you can download from CPAN, provides
1526 many ready-to-use subroutines that you can use with L<File::Find>.
1528 The L<File::Finder>, which you can download from CPAN, can help you
1529 create the callback subroutine using something closer to the syntax of
1530 the C<find> command-line utility:
1535 my $deep_dirs = File::Finder->depth->type('d')->ls->exec('rmdir','{}');
1537 find( $deep_dirs->as_options, @places );
1539 The L<File::Find::Rule> module, which you can download from CPAN, has
1540 a similar interface, but does the traversal for you too:
1542 use File::Find::Rule;
1544 my @files = File::Find::Rule->file()
1548 =head2 How do I delete a directory tree?
1550 (contributed by brian d foy)
1552 If you have an empty directory, you can use Perl's built-in C<rmdir>.
1553 If the directory is not empty (so, no files or subdirectories), you
1554 either have to empty it yourself (a lot of work) or use a module to
1557 The L<File::Path> module, which comes with Perl, has a C<remove_tree>
1558 which can take care of all of the hard work for you:
1560 use File::Path qw(remove_tree);
1562 remove_tree( @directories );
1564 The L<File::Path> module also has a legacy interface to the older
1565 C<rmtree> subroutine.
1567 =head2 How do I copy an entire directory?
1569 (contributed by Shlomi Fish)
1571 To do the equivalent of C<cp -R> (i.e. copy an entire directory tree
1572 recursively) in portable Perl, you'll either need to write something yourself
1573 or find a good CPAN module such as L<File::Copy::Recursive>.
1575 =head1 AUTHOR AND COPYRIGHT
1577 Copyright (c) 1997-2010 Tom Christiansen, Nathan Torkington, and
1578 other authors as noted. All rights reserved.
1580 This documentation is free; you can redistribute it and/or modify it
1581 under the same terms as Perl itself.
1583 Irrespective of its distribution, all code examples here are in the public
1584 domain. You are permitted and encouraged to use this code and any
1585 derivatives thereof in your own programs for fun or for profit as you
1586 see fit. A simple comment in the code giving credit to the FAQ would
1587 be courteous but is not required.