#!/usr/bin/perl
use NetAddr::IP;  #used for CIDR
use Getopt::Long; # handle command line arguments/flags
use File::Basename;
# use strict;      # Okay, I'm gonna take some heat for this.
# use warnings 'all';  # Yeah, this too.
use Text::CSV qw( csv ); # see https://metacpan.org/pod/Text::sort { $
+a <=> $b } keys %hash;CSV for info
use open ":std", ":encoding(UTF-8)";
use Data::Dumper; # use print Dumper($scalar_var); to see data structu
+res or objects

$command = basename($0);

sub help {
print "\n";
print "perl $command -r report# report_name     Use perl $command -l t
+o list supported report formats\n\n";
print "example: perl $command -r 4b May24.csv\n\n";
print "\nUSAGE:\n";
print "               -h -H or -help         Display this help\n";
print "               -r report#             Required argument - repor
+t number\n";
print "               -l -L or -list         List supported report num
+bers\n";
print "               report                 Report name to process (c
+sv format)\n";
exit;
} # sub help

sub list {
    print "Supported report formats:  ". join(" ",@supported_reports) 
+. " \n";
    print "    4b - User or Group Permissions for Directory - (compres
+s report based on folders and permissions)\n\n";
    print "\n";
    help;
    exit;
} # sub list


# Default values for options
# $list = ""; 
# $help = "";

# GetOptions complains if you use upper AND lower case options below. 
+It converts args into lowercase.
GetOptions (
    'list|l' => sub { list },
    'report|r=s' => \$report,
    'help|h' => sub { help }
);

# Put supported report formats here - note the format!
# @supported_reports = qw( 4b 5c 6d); # multiple report formats can be
+ listed like this - need subroutines for each
@supported_reports = qw( 4b );

# list if $list;
help if ($#ARGV >= 1); #needs minimum -r <reportnum> filename as argum
+ents - should be nothing but a filename now

# assign filename
($filename) = @ARGV; # (should only have one argument)

# see if report is supported - %reports is a hash array
unless ( grep {$_ eq $report } @supported_reports ) {
  print "Report format $report is NOT supported\n";
  list();
  exit(1);
}

# can't get map to work except with @ARGV - don't know why - keeping f
+or later analysis
#my %reports = map { $_ => 1 } @supported_reports;
#if (exists($reports{$report})) { ... }


# I started to work with manually parsing them but ran into embbeded c
+ommas in fields, and found Text::CSV.  It slurps in everything, but p
+ulls it into a hash of hashes (never worked with that before, but tho
+ught I'd try to slog through it as a learning exercise.
# Open file, if it exists
open(my $fh , '<', $filename) or die "Unable to open file, $filename";
# old method slurping in lines manually
# Grab 1st line headers
$first_line = <$fh>;
$first_line =~ s/\x{FEFF}//;  # strip out leading UTF8 Byte Order Mark
+er (BOM)
chomp $first_line; # strip off trailing carriage return
@headers = split(",",$first_line);  # Grab headers into array, split o
+n commas
$num_headers = scalar @headers; # get number of headers.

# Here's the newer stuff
# $aoa and $aoh are hash refs (in a scalar) - need to figure out how t
+o parse/manipulate them.  And build new ones
$aoa = csv (in => $filename , encoding => "UTF-8"); # this slurps a fi
+le into an array of arrays (need to find a way to sort this and also 
+remove all the UTF-8 encodings on each line.)
$aoh = csv (in => $filename , headers => "auto", encoding => "UTF-8");
+ # this slurps a file into an array of hashes (need to find a way to 
+sort this and also remove all the UTF-8 encodings on each line.)
# pre-process keys to remove unicode embedded in key names - Found in 
+first few characters of 1st line, but was embedded in all the arrays.
+ (this fixed it!)
for $key ( keys @{$aoh} ) {
  $data = $aoh->[$key];
  for $temp_key (keys %$data) {
      if ($temp_key =~ /\x{FEFF}/) {
          $new_key = $temp_key;
      $new_key =~ s/\x{FEFF}//;
      $data->{$new_key} = delete $data->{$temp_key};
      $temp_key = $new_key;
          print "------> Data key $data->{$temp_key} \n"; #print state
+ments to look at data
      }
  print "Temp key $temp_key \n"; #print statements to look at data
  print "Data key $data->{$temp_key} \n"; #print statements to look at
+ data
  }
}  

# These were earlier attempts to understand the data structure - I can
+ delete this
for $key ( keys @{$aoh} ) {
  $data = $aoh->[$key];
  print "key: $key, data $data \n";
  print "key: $key, values:\n";
  for $field ( keys %$data ) {
    $temp_field = $field;
#    $field =~ s/\x{FEFF}//;  # (error in read only field)
    print " temp_field = $temp_field    field = $field \n";
    printf "    %-18s => %s\n", $field, $data->{$field};
  }
}

# for $key ( keys @{$aoh} ) {
#   $data = $aoh->[$key];
#   print "key: $key, values:\n";
#   for $field ( "Logon Name", "Current Permissions" ) {
#     printf "    %-18s => %s\n", $field, $data->{$field};
#   }
# }
# Generated
# key: 0, values:
#    Logon Name         => @FOO NOW Onsite Support
#    Current Permissions => FMRWX
# key: 1, values:
#    Logon Name         => Administrators
#    Current Permissions => FMRWX
# key: 2, values:
#    Logon Name         => Creator Owner
#    Current Permissions => FMRWX


# route errors to the bit bucket (disappears)
open STDERR, '>/dev/null';


# For each server and Access Path (file) basis, Collect the Current Pe
+rmissions, and concatenate all the logon names with that same permiss
+ion.  Note when it's not inherited "<not inherited>" in the Inherited
+ From Folders field.
#
# {File Server}{Access Path}{Current Permissions}{Logon Name}{Inherite
+d From Folders}{Flags}{User/Group}{Classification Results}.....
#
# 4b headers
# 0  'File Server'
# 1  'Access Path'
# 2  'Current Permissions'
# 3  'Logon Name'
# 4  'Inherited From Folders'
# 5  'Flags'
# 6  'User/Group'
# 7  'Classification Results'
# 8  'Classification Results by Category (Including Nested)'
# 9  'Total Hit Count'

# Process 4b report
# Need code about dynamic arrays here.

# Early code to manually loop through the file, but found embedded com
+mas in various fields, throwing the count of fields relative to the h
+eaders. Gave it up after finding TEXT::CSV 
# Loop through file
$line_count = 0;
while (<$fh>) {
    $line_count++;
    chomp; # remove CR/LF
    @curr_line = split(",",$_);
    print "Line $line_count Field count != header count - Could be ext
+ra commas embedded\n  $_ \n\n" if ((scalar @curr_line) != (scalar @he
+aders));
    exit if ($line_count == 10);    
} # (<$fh>)
[download]