# concordance - create a list of all places in a set of files where a given
# set of words occurs

use MakeRegex;

my @list= qw(arm back leg elbow ear eye chin cheek knee head hip tooth teeth lips mouth hair);
   $dir = "c:\\temp\\";
concordance($dir,\@list);
exit;


sub concordance {
   ($dir,$list) = @_;

   # makeregex from word list
   my($regex) = MakeRegex::make_regex(@$list);
   print "regex: $regex\n";

   opendir(DIR,$dir) or die "Can't open $dir: $! \n";

   #iterate over all the files in a directory
   @files = readdir(DIR);
   foreach $file (@files) {
      if (-f "$dir\\$file") { 
         print "infile: $dir\\$file\n"; 
         open(INFILE,"$dir\\$file") or die "Can't open file $file for input\n";

         # iterate over the lines of the file
         $line_no = 1;
         while($line = <INFILE>) {
            if ($line =~ /($regex)/) {
               $word = $1; # first word matched in a line
               print "($word,$file,$line_no): $line\n";
            } 
            $line_no++;
         }
         close(INFILE);
      }
   }
   closedir(DIR);
}

# next:
# invertfile - create an index for a set of keywords in a set of files.
# (word;line;para,char)
# can't smush all lines of a file into one gigantic string because 
# the files that hold literary works can often be quite big.