#!perl

=head2 NAME

Find regex in field extract

=head2

Use as template for finding a regular expression in individual fields.
Input file should be result (export file) from fieldextraction.pl or similar file.
Output file is list of lines matching the desired regular expression.

=cut 

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

open (IN, "<$inputfile") || die ("can't open in");
open (OUT, ">$exportfile") || die ("can't open out");
#if using MacPerl, set creator and type to BBEdit and Text
if ($^O eq 'MacOS') {
MacPerl::SetFileInfo('R*ch', 'TEXT', $exportfile);
}

my $linecount = 0;
while (my $line = <IN>) {
chomp $line;
#remove extra spaces between tabs
$line =~ s/\t\s{7}\@/\t\@/g;
#remove count, tag, and first subfield code
$line =~ s/^\d+?\t\d{3}\s[ \d][ \d]\s\@\w\t//;

#change to 'if' for opposite
#change text between // for desired regex
unless ($line =~ /\t/) {
print OUT "$line\n";
$linecount++;
}
}
close IN;
close OUT;
print "$linecount lines identified\n";
print "Press Enter to quit";
<>;

=head1 LICENSE

This code may be distributed under the same terms as Perl itself. 

Please note that this code is not a product of or supported by the 
employers of the various contributors to the code.

=head1 AUTHOR

Bryan Baldus
eija [at] inwave [dot] com

Copyright (c) 2003-2004

=cut