Hello,

I am playing around with my NAS and have this qpkg package that I want to split on my Windows machine. The qpkg package is just too much to post here but roughly it consists of:

...sh script...
something.tar.gzNULNULNUL...bunch tarball here....7z¼¯' ...bunch of 
+7z compressed material here...NULNULBS=MIME-Version: 1.0 ... base64 e
+ncoded material here...
[download]

The sh script has lines:

script_len=3265
/bin/dd if="${0}" bs=$script_len skip=1 | /bin/tar -xO | /bin/tar -xzv
+ -C $_EXTRACT_DIR || exit 1
offset=$(/usr/bin/expr $script_len + 20480)
/bin/dd if="${0}" bs=$offset skip=1 | /bin/cat | /bin/dd bs=1024 count
+=26 of=$_EXTRACT_DIR/data.tar.7z || exit 1
[ -f /usr/local/bin/python ] && /usr/local/bin/python -c "with open('$
+_EXTRACT_DIR/data.tar.7z', 'rw+') as f: f.seek(26585); f.truncate()"
offset=$(/usr/bin/expr $offset + 26585)
[download]

From that I calculated the following start positions:

0, 3265, 3265+20480, 3265+20480+26585

So I wrote a quick script to extract this stuff, but now I notice substr giving me trouble, most of the time it seems to output too many characters. But the start positions seems to be ok. I don't understand why. Any ideas?

#!/usr/bin/perl

use strict ;
use warnings ;
use File::Basename ;

# binmode(STDOUT, ":utf8");

my $fn = $ARGV[0] ;
( my $fns, my $fp, my $fs ) = fileparse( $fn, qr/\.[^.]*/ ) ;
# print "fns = $fns, fp = $fp, fs = $fs\n" ; 
my $fc ;
{
    local $/ ; # slurp
    # ? '<:encoding(utf8)'
    # open (my $fh, '<:encoding(utf8)', $fn ) or die "Failed to open f
+ile $fn with error: $!\n" ;
    open (my $fh, '<', $fn ) or die "Failed to open file $fn with erro
+r: $!\n" ;
    binmode $fh ; # Edit suggestion Anonymous Monk on Feb 14, 2021 at 
+21:31 UTC
    $fc = <$fh> ;
    close $fh ;
}
my $number = 'x' ;
my @numbers ;
while ( $number ne '' ) {
    print "Type number where to split: " ;
    $number = <STDIN> ;
    chomp $number ;
    push( @numbers, $number ) unless $number eq '' ;
}
my $start = 0 ;
for my $i (0 .. $#numbers) {
    print "$i: $numbers[$i]\n" ;
    my $fn2 = $fp . $fns . "." . ($i+1) ;
    open (my $fh, ">", $fn2 ) or die "Failed to open file $fn2\n" ;
    binmode $fh ; # Edit suggestion Anonymous Monk on Feb 14, 2021 at 
+21:31 UTC
    print $fh substr($fc, $start, $numbers[$i]) ;
    $start = $numbers[$i] ;
    close $fh ;
}
[download]

I have tried to set file input and stdout to utf8 (commented lines) but that makes things worse