beer planet » symlink http://beerpla.net where things have nothing to do with beer - tutorials, tips, how-tos, thoughts, hacks, and other techy nonsense Fri, 06 Jan 2012 08:50:59 +0000 en hourly 1 http://wordpress.org/?v=3.3.1 [Perl] Finding Files, The Fun And Elegant Wayhttp://beerpla.net/2009/04/08/perl-finding-files-the-fun-and-elegant-way/ http://beerpla.net/2009/04/08/perl-finding-files-the-fun-and-elegant-way/#comments Wed, 08 Apr 2009 14:00:00 +0000 Artem Russakovskii http://beerpla.net/2009/04/08/perl-finding-files-the-fun-and-elegant-way/ No matter what programming language you use, there comes a time when you need to search for a file somewhere on the file system. Here, I want to talk about accomplishing this task in Perl. There are many ways of doing so, most of them boring, but I want to discuss the fun and elegant way – using File::Find::Rule.

Let me briefly discuss some of the other methods first.

Limited

Using glob() (or <>, TODO verify) you can find files in a single directory, using only the limited shell wildcard support. For example,

1

my @files = glob("tmp*");

I prefer glob() to <> because glob()'s parameters can be more than just text (for ex functions) while <>
...
Read the rest of this article »

]]> No matter what programming language you use, there comes a time when you need to search for a file somewhere on the file system. Here, I want to talk about accomplishing this task in Perl. There are many ways of doing so, most of them boring, but I want to discuss the fun and elegant way – using File::Find::Rule.

Let me briefly discuss some of the other methods first.

Limited

Using glob() (or <>, TODO verify) you can find files in a single directory, using only the limited shell wildcard support. For example,

1

my @files = glob("tmp*");

I prefer glob() to <> because glob()'s parameters can be more than just text (for ex functions) while <> treats everything inside as text.

Boring

File::Find is the de facto standard for searching in Perl.

This method finds files that end in .pl in "." and "../SomeDir", following symlinks:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21

#!/usr/bin/perl -w
 
use File::Find;
use Data::Dumper;
use File::Basename;
my @directories_to_search = (".", "../SomeDir");
my @file_list = ();
 
find(
  { wanted =>
    sub {
      if ( basename($File::Find::name) =~ /\.pl$/i )
      {
        push @file_list, $File::Find::name;
      }
    },
    follow => 1
  },
  @directories_to_search
);
print Dumper @file_list;

It works fine, except it's horribly ugly and boring. Let's have a look at something more fun.

The Fun And Elegant Way

File::Find::Rule. Just have a look at this beauty.

Just like above, find all .pl files in "." and "../SomeDir", following symlinks:

1
2

print Dumper (File::Find::Rule->name("*.pl")->file->extras({ follow => 1 })->
in(".", "../SomeDir"));

Same as above, except bypass .svn directories (shaves off a ton of time with a lot of directories):

1
2

print Dumper (File::Find::Rule->not(File::Find::Rule->directory->name('.svn')->
prune->discard)->name("*.pl")->file->extras({ follow => 1 })->in(".", "../SomeDir"));

Find all .log files that are older than 24 hours in "."

1
2
3

my $epoch_time_1_day_ago = time() - 60*60*24;
print Dumper (File::Find::Rule->file->name("*.log")->
mtime("<$epoch_time_1_day_ago")->in('.'));

Be sure to read the File::Find::Rule perldoc for more options and remember: have fun with your code!

Thanks to Perlbuzz and Andy Lester for pointing me to this library a few months ago.

Similar Posts:

Share

]]> http://beerpla.net/2009/04/08/perl-finding-files-the-fun-and-elegant-way/feed/ 2