bin/image-db-build.pl

#!/usr/bin/perl

# create a list of image files, mtime, and sha256sum for later analysis

use v5.12;    # 5.12+ gives strict and say...
use warnings;
use File::Find;
use Digest::SHA;
use Image::Info qw(image_type);
use Term::ProgressBar::Simple;
use Text::CSV;

my @search_dirs = @ARGV or die "Need at least one directory name:$!";

my $csv = Text::CSV->new( { binary => 1, eol => qq{\r\n} } )
  or die "Cannot use CSV: " . Text::CSV->error_diag();

my $dbname = q{fileinfo.db};
open my $FH, q{>:encoding(utf8)}, $dbname or die "$dbname: $!";

my $tot_files = 0;
find( sub { $tot_files++ if -f $File::Find::name }, @search_dirs );

say "Scanning $tot_files files in "
  . join( q{ }, @search_dirs )
  . ".  Saving data to $dbname.";

my $progress = Term::ProgressBar::Simple->new($tot_files);
find( \&process_file, @search_dirs );

sub process_file {

    return if not -f $File::Find::name;

    $progress++;

    if ( not image_type($File::Find::name)->{error} ) {

        my $sha = Digest::SHA->new(q{sha256});
        $sha->addfile( $File::Find::name, q{b} );

        my (
            $dev, $ino,  $mode, $nlink, $uid,
            $gid, $rdev, $size, $atime, $mtime
        ) = stat($File::Find::name);

        $csv->print( $FH,
            [ $File::Find::name, $_, $mtime, $size, $sha->hexdigest ] );
    }    # if it's an image
}    # sub process_file

__END__