Re: Strange output problem

[Lotus1]

I noticed that you re-open the same file from $out_file as $new_in_file without closing $out_file first. Seems like buffering could be an issue here.

I tried your program with a smaller number of rows and columns and found that with the close statement before the reopen I got output in the final processed file and without it I got a blank file. The intermediary file looked good in both cases. Maybe the final data to the intermediary gets written when the program ends and that first file handle is closed. I don't want to troubleshoot much here until you have a look at using close.

Update: I thought of an easy way to test my idea about buffering. I made a copy of the intermediary file then closed the intermediary file then made another copy. The first copy was empty the second had everything in it.

# Remove unnecessary rows, clean up the SNPs and put into CSV format:

open my $new_in_file, '<', "chr${chr_num}_exome_snps_only_transposed_$
+cont"
  or die "Cannot open new input file: $!";

##################################
#    I added the following lines 
#  
open my $temp, '>', 'transposed_copy' 
    or die "Cannot open transposed_copy: $!";  

while(<$new_in_file>) {
    print $temp $_;
}
close $temp;
    
close $out_file;

open $temp, '>', 'transposed_copy2' 
    or die "Cannot open transposed_copy2: $!";  

while(<$new_in_file>) {
    print $temp $_;
}
close $temp;