Re: Finding files recursively

[holli]

You are fighting the module. And you are doing a lot of unneccessary work. Consider

use File::Find;

my @found;
my $path   = 'd:\env\videos';
my $target = '2012.avi';

find ( sub 
{ 
    # We're only interested in directories
    return 
        unless -d $_;
    
    # Bail if there is an .ignore here
    return 
        if -e "$_/.ignore";
    
    # Add to the results if the target is found here
    push @found, $File::Find::name
        if -e "$_/$target";

}, $path);

print "@found";

D:\ENV>perl pm10.pl
d:\env\videos/2012
D:\ENV>echo.>d:\env\videos\2012\.ignore
D:\ENV>perl pm10.pl

D:\ENV>

[ovedpo15]

Thanks for your suggestion but I don't understand the difference between both suggestions. Also, what is $target?. Thank you again.

[holli]

$target is just the filename you are looking for, "secret.file" in your case.
The difference is that my code is exiting the wanted function immedeatly when it is not dealing with a directory. Only if there is a directory it is looking wether the target file is in that directory.

Whereas your code looks at each and every file, calculates its' base path (albeit unneccessary, that info is already there in $File::Find::name). And then it takes that base directory to look for the target file.
This, and this is the biggest slowdown, also means that you are testing the same directory number-of-entries-in-the-directory times.

[ovedpo15]

Back with results! :)
I tried my code and your code. My code ran for 13858 seconds and your code ran for 16968 seconds. I thought it will reduce the time a little but it didn't, maybe because the machine was being used by others at that time but it is a big difference. Do you have any other suggestions? 4 hours for searching is quite a lot of time :(