#!/usr/bin/perl

use strict;

my ($xml_source, $data_source, $filter_args) = @ARGV;

#print "$xml_source\n";
#print "$data_source\n";

#print "$filter_args\n";
my @filter_args = split (/\,/, $filter_args);

#call data args into scope from xml document

use LWP::Simple;
my $content;

$content = get $xml_source;
die "Couldn't get $xml_source" unless defined $content;

use XML::Simple;
my $xs1 = XML::Simple->new();
my $doc = $xs1->XMLin($content, forcearray => 1);

=comment
#to view the contents of the $doc array
use Data::Dumper;
print Dumper($doc);
=cut

my ($file_row_start, $file_row_comment, $file_field_separator, $file_field_missing_value, $column_time) ;

my $file_row_start = $doc->{'global_attributes'}->[0]->{'file_row_start'}->[0] ;
if ($file_row_start =~ /HASH/) { $file_row_start = 1; }

#assumes 'none' won't actually be used as a line comment
#not testing for escape characters
my $file_row_comment = $doc->{'global_attributes'}->[0]->{'file_row_comment'}->[0] ;
if ($file_row_comment =~ /HASH/) { $file_row_comment = 'none'; }

#not testing for escape characters
my $file_field_separator = $doc->{'global_attributes'}->[0]->{'file_field_separator'}->[0] ;

#assumes 'none' won't actually be used as a missing value
#not testing for escape characters
my $file_field_missing_value = $doc->{'global_attributes'}->[0]->{'file_field_missing_value'}->[0] ;
if ($file_field_missing_value =~ /HASH/) { $file_field_missing_value = 'none'; }

my $column_time = $doc->{'global_attributes'}->[0]->{'column_time'}->[0] ;
if ($column_time =~ /HASH/) { $column_time = 1; }

my $measurement_time_zone = $doc->{'global_attributes'}->[0]->{'measurement_time_zone'}->[0] ;
if ($measurement_time_zone =~ /HASH/) { $measurement_time_zone = '+00'; }

#populate local array from ascii column data using xml file processing args

$content = get $data_source;
die "Couldn't get $data_source" unless defined $content;

#split rows into array elements
my @data_content = split (/\n/, $content);

my @obs_values;
my $number_file_columns;

my $line;
my @line_array;
my $line_count_file = 0;
my $line_count_processed = 0;

foreach $line (@data_content) {
        $line_count_file++;

        #drop lines until first row start
        if ($line_count_file < $file_row_start) { next; }

	#skip lines which are commented
        if ($line =~ /^$file_row_comment/) { next; }

	#processing row
	#print "process\n";
        $line_count_processed++;

	@line_array = split(/$file_field_separator/,$line);

	#map missing values
	#foreach my $element (@line_array) {
	#	if ($element eq $file_field_missing_value) { $element = '-99999'; }
	#}	

	push (@obs_values, @line_array);
}

my $number_file_columns = scalar(@line_array);
#print "line_count_processed: $line_count_processed\n";

#now process populated array 
process_array_column(\@obs_values, $column_time); 

my $filename = 'filter_'.int(rand(10000000)); 

#create temp file
open (TEMP_FILE, ">/tmp/ms_tmp/$filename\.txt");
print TEMP_FILE print_array(\@obs_values);
close (TEMP_FILE);

#create temp file ref
open (TEMP_FILE, ">/tmp/ms_tmp/$filename\_ref.txt");
print TEMP_FILE "http://nautilus.baruch.sc.edu/ms_tmp/$filename\.txt";
close (TEMP_FILE);

#the filename handle is passed back to the calling php page for reference
print $filename; 

exit 0;

sub process_array_column {

my ($array_line, $get_column) = @_;

my $column_count = 1;
#print $measurement_time_zone;

foreach my $element (@{$array_line}) {
	if ($column_count == $get_column) {
		if ($filter_args[0] eq 'breakoutYYYYMMDDHHMM') {
		        $element = substr($element,0,4).'-'.substr($element,4,2).'-'.substr($element,6,2).' '.substr($element,8,2).':'.substr($element,10,2).':00'.$measurement_time_zone.'00';
		}
		if ($filter_args[0] eq 'breakoutNormal') {
		        $element = substr($element,0,19).$measurement_time_zone.'00';
		}
	}
	
	$column_count++;
	if ($column_count == $number_file_columns+1) { $column_count = 1; }
}

} #process_array_column

sub print_array {

my ($array_line) = @_;

my $column_count = 1;
my $array_list = '';

foreach my $element (@{$array_line}) {
        $array_list .= $element;
        #if ($column_count != $number_file_columns) { $array_list .= field_separator_restore($file_field_separator); }
        if ($column_count != $number_file_columns) { $array_list .= "_SEP_"; }

        $column_count++;
        if ($column_count == $number_file_columns+1) { $column_count = 1; $array_list .= "\n"; }
}

return $array_list;

} #print_array

sub field_separator_restore {

my ($field_separator) = @_;

if ($field_separator eq '\,') { return ","; }
if ($field_separator eq '\t') { return "\t"; }
if ($field_separator eq '\s+') { return " "; }

return $field_separator;
}