#!/usr/local/bin/perl

use strict;

if ($#ARGV < 1) {usage();}

my $logfile = $ARGV[0];
my $rptfile = $ARGV[1];

# setup output
my $fopen = sprintf("Cannot open %s for output\n", $rptfile);
open(rptfile, ">$rptfile") or die $fopen;

# get data
$fopen = sprintf("Cannot open %s for input\n", $logfile);
open(logfile, $logfile) or die $fopen;
my @summdat = <logfile>;
close(logfile);
shift @summdat;   # remove unnecessary first line

foreach my $summline (@summdat)
{
  my ($junk, $id, @rest) = split /:/, $summline;

  $id =~ tr/a-zA-Z0-9/ /c;   # remove punctuation
  $id =~ s/^ +//;            # remove leading space
  my ($ida, $idb) = split / /, $id;
  $id = sprintf ("%-4.4s %10.10s:", $ida, $idb);

  my $what = '';
  foreach my $item (@rest) {$what .= ':' . $item;}
  $what =~ s/^\://;

  next if $what =~ ' already in unicode';

  if ($what =~ / Error.+/)
  {
    my $stop = index($what, "'");
    printf rptfile ("%s   %s\n", $id, substr($what, 0, $stop-1));
  }
  else
  {
    my ($number, $data) = split /:/, $what;
    $number =~ s/[\(\)\[\]]/ /g;   # remove ()[]
    $number =~ s/\s\s*/ /g;        # collapse spaces down to one
    $number =~ s/^ +//;
    my ($fieldn, $whichfield) = split / /, $number;

    my $where = index($data, "at") + 3;
    my $at = '';
    my $char = substr($data, $where++, 1);
    while ($char =~ /\d/g)
    {
      $at .= $char;
      $char = substr($data, $where++, 1);
    }

    my $here = length($data) - 2;   # before last (')
    my $somechars = substr($data, $here-8, 8);    # last 8 chars

    printf rptfile "$id   ";
    printf rptfile ("field # %s, which is tagged %s, around char # %s",
                    $fieldn, $whichfield, $at);
    if ($somechars ne '........')
      {printf rptfile (", something like '%s'\n", $somechars);}
    else
      {print rptfile "\n";}
  }
}
close(rptfile);


sub usage()
{
  print ("\nUsage: ulogtranslate.pl summaryfile rptfile\n\n");
  print ("   Reads the specified summaryfile that resulted from\n");
  print ("   the Voyager Unicode conversion.\n");
  print ("   A human readable translation of this data is put in rptfile.\n");
  print ("   Include the file path if needed. For example, for summaryfile\n");
  print ("   it might be: /m1/incoming/2005.0/INFO/dbname/summary.txt\n");
  print ("   where you substitute your database name of choice for \"dbname\".\n\n");
  print ("   [This program is designed to read the summary.txt file\n");
  print ("   that the Unicode conversion process creates.]\n\n");
  exit(0);
}