#!perl

=head2 

Identify miscodings difference.

Use to identify bad or missing headings when comparing two similar files.

Reads in base file, containing headings to compare against.
Reads in lines from extractedfile, looks for match in base file.
If match is not found, outputs the extractedfile line to new file for manual cleanup.

=cut 

print "File of base headings: ";
my $baseinputfile = <>;
chomp $baseinputfile;
$baseinputfile =~ s/^\"(.*)\"$/$1/;

print "Input file to clean: ";
my $inputfile  = <>;
chomp $inputfile;
$inputfile =~ s/^\"(.*)\"$/$1/;

print "Export file: ";
my $exportfile = <>;
chomp $exportfile;
$exportfile =~ s/^\"(.*)\"$/$1/;

open (BASEHDGS, "<$baseinputfile") || die ("can't open base");
my @baseheadings = ();

while (my $baseline = <BASEHDGS>) {
chomp $baseline;
#remove ending periods and trailing spaces
#$baseline =~ /\.\s*$/;
push @baseheadings, $baseline;
} # end reading base file

open (IN, "<$inputfile") || die ("can't open in");
open (OUT, ">$exportfile") || die ("can't open out");
my $linecount = 0;
while (my $line = <IN>) {
chomp $line;
#remove extra spaces between tabs
$line =~ s/\t\s{7}\@/\t\@/g;

#remove ending periods and trailing spaces
#remove count, tag, indicators, and first subfield code and char
$line =~ s/^\d+?\t\d{3}\s[ \d][ \d]\s\@\w\t//;

#split control nos from heading if subfields exist or control numbers are present
if (($line =~ /\t\@\w\t/) || ($line =~ /\t\t/)) {
my @linearray = split ("\t\t", $line);
#break subfields into individual slots
my @headingarray = split ("\t", $linearray[0]);
#look at first subfield
my $lineheading = $headingarray[0];
$lineheading =~ /\.\s*$/;
}
else {$lineheading = $line;} #controlfield extracts
my @grepheadings = grep {$_ eq $lineheading} @baseheadings;

unless (@grepheadings) {
print OUT "$line\n";
$linecount++;
}
}
close BASEHDGS;
close IN;
close OUT;
print "$linecount lines identified\n";
print "Press Enter to quit";
<>;