Beefy Boxes and Bandwidth Generously Provided by pair Networks
Welcome to the Monastery

Answer: How do I search a directory tree for files?

by QandAEditors
on Jun 28, 2000 at 12:28 UTC ( #20136=categorized answer: print w/replies, xml ) Need Help??

Q&A > directories > How do I search a directory tree for files?

There are two possibilities, depending on how static the directory contents are, and how willing you are to trade speed against memory.

The first solution searches the whole tree. This is the solution to go for if the directories themselves are static, but the contents of the directories are not static. This version is slow, but it dosen't consume much memory (on harddisk).

#!/usr/bin/perl -w use strict; use File::Find; my @directories = (".", "/home/mp3"); my @foundfiles; # Here, we collect all .mp3 files below each directory in @directories + and put them # into @foundfiles find( sub { push @foundfiles, $File::Find::name if /\.mp3$/ }, @direct +ories ); # and output them all print join("\n",@foundfiles), "\n";

The second version uses a two step approach. We compute a list of all (interesting) files in the directory tree once, and save it into a file. If we want to check if a certain file is in the directory tree, we load this file into a hash and have a really fast lookup (if we want to look up more than one file) or we go through the file line by line (if we only look for a single file). This method obviously only works if the directory contents don't change very often, because our file is not always up-to-date. The code above serves very well to create the list of interesting files, just redirect its output into a file called index.

#!/usr/bin/perl -w use strict; use File::Basename; my %files; my @searchfiles = ("foo", "bar", "xxx"); open( INDEX, "< index" ) or die "Couldn't read index : $!\n"; # now we read every filename from our index file and put it in the has +h. # If we are only checking for one file, we could do the check right he +re # in the loop. # We also strip the path from the filename, as we will be searching fo +r files # (and if we already knew the path to the file, -e would be faster :) +) # This method dosen't care for duplicates. If we have two files with t +he same name, # only the last file will be reported. my $filename; while( <INDEX> ) { $files{ basename( $_ ) } = $_; }; close INDEX; # And now we check if the filenames are in the hash foreach (@searchfiles) { print $files{ $_ } if (exists $files{ $_ }); };

Log In?

What's my password?
Create A New User
and the web crawler heard nothing...

How do I use this? | Other CB clients
Other Users?
Others pondering the Monastery: (6)
As of 2018-11-13 03:14 GMT
Find Nodes?
    Voting Booth?
    My code is most likely broken because:

    Results (149 votes). Check out past polls.