#!/usr/bin/perl
#############################################################################
# PERL Program "fnaprocess.pl
# 
# April 8th, 1998
#
# AUTHORS
# Anton Chuppin
# Eric Larson
#
# PURPOSE
# This program processes the fnaindex files into a usable form.
#
###################################################################
 
$numargs = @ARGV ;

if ($numargs ne "2") {
	print "Usage: ", $0, "\"Input sub-dir\" \"Output sub-dir\" -- Use trailing backslash\n" ;
	die ;
}

$subdir = $ARGV[0] ;    # Input sub-dir name
$outdir = $ARGV[1] ;    # Output sub-dir name

unless (((-d $outdir)&&(-w $outdir)) || mkdir($outdir,0775)) {
    die ("$0 died -- output directory $outdir could not be created 
        or written to\n");
}

unless (opendir (DATADIR, $subdir)) { ;    # Open up the data sub-directory
    die "$0 Died -- $subdir data input sub-dir cannot be opened" ;
}

while ($filelist = readdir(DATADIR)) {         # Read the filelist in
  $readfile = $subdir . $filelist ;            # Create the input file path
  $skipfile = 0 ;
  unless (open(INFILE, "$readfile")) {         # Open the input file
     print STDERR "$0 Error: ".
        "$filelist input file cannot be opened -- skipping\n" ;
     $skipfile = 1 ;
  }
  if ($skipfile = 1 ) {
    $outfile = $outdir . $filelist ;             # Create the output file path
     unless (open (OUTFILE, ">>$outfile")) {      # Open the output file
       if (!(-w $outfile)) {
         warn ("$0 error -- No write permissions to $outfile \n") ;
       } else {
         warn ("$0 error -- Outfile $outfile cannot be opened \n") ;
       }
     }
    while ($line = <INFILE>) {
      chomp($line) ;
      $line =~ tr/A-Z/a-z/;                 # Convert to lowercase
      $line =~ tr/-/_/ ;                    # Convert dashes to underlines
      $line =~ s/\&nbsp\;/_/ ;              # Find &nbsp; and convert to _
      $line =~ s/\s+/ /g ;                  # Remove excess spaces 
      $line =~ s/\<b\>\s+/\<b\>/g ;          # Remove spaces after <b>
      $line =~ s/\s+\<\/b\>/\<\/b\>/g ;      # Remove space before </b>
      $line =~ s/\<b\>/\n\<b\>/g ;        # Add newlines before <b>
      $line =~ s/\<b\>(.*?)\<\/b\>/{$tmp=$1;$tmp=~s|\s+|_|g;$tmp;}/gie;
      # $line =~ s/\<b\>(.*?)\s(.*?)\<\/b\>/\1_\2/g ; # Trans fourth bold space
      $line =~ s/\<.*?\>/ /g ;              # Remove HTML, add space
      $line =~ s/ +/ /g ;                   # Remove excess spaces 
      $line =~ s/\n /\n/g ;
      
    print OUTFILE $line ;
    }
    close (INFILE) ;                          # Close the current input file
    print OUTFILE "\n";
    close <OUTFILE> ;                         # Close the current input file 
  }                                           # End the "skipfile" block
}                                             # End the fileread block  

print ("$0 terminated normally\n") ;
exit (0);