CSCI 3333
Data Structures
Summer 2008
Homework #3 Possible Solution

The following is a possible solution in Perl, since I do not have time to develop Java/C++ solutions.

#!/opt/gnu/bin/perl

#  Create Jumbo games: part #1.
#  Create files of words made of unique set of characters.
#  All unique words of length 5 or below go to the file word_set1.txt
#  All unique words of length 5 or above go to the file word_set2.txt
#  All non-unique words go to the file other_word_set.txt

open(DICT, "words.txt") || die "Can't open words.txt";
while ($word = <DICT>)
{  chop($word);
   #  remove single-character words, capitalized words
   #  and words with a '.
   next if ($word =~ /^.$/ || $word =~ /^[0-9A-Z]/ || $word =~ /'/);
   $key = join(//,sort(split(//, $word)));

   if ($chset{$key})
   {   if ($chset{$key} != -1) {
           # word is not unique. Store the first word with the same key in @nonuniqueSet.
           push(@nonuniqueSet, $chset{$key});
           $chset{$key} = -1;
       }
       # Store the current word with the same key in @nonuniqueSet.
       push(@nonuniqueSet, $word);
   }
   else
   {  $chset{$key} = $word;
   }
}
close(DICT);

foreach $key (keys %chset)
{  if ($chset{$key} == -1) {
       next;
   }
   if (length($key) >= 5)
   {  push(@word_set2, $chset{$key});
   }
   else
   {  push(@word_set1, $chset{$key});
   }
}

@nonuniqueSet = sort(@nonuniqueSet);
open(OTHER_WORD_SET, ">other_word_set.txt")  || die "Can't open other_word_set.txt" ;
while (@chset)
{  $_ = shift(@nonuniqueSet);
   print OTHER_WORD_SET "$_\n";
}
close(OTHER_WORD_SET);

@word_set1 = sort(@word_set1);
open(WORD_SET1, ">word_set1.txt") || die "Can't open word_set1.txt" ;
while (@word_set1)
{  $_ = shift(@word_set1);
   print WORD_SET1 "$_\n";
}
close(WORD_SET1);

@word_set2 = sort(@word_set2);
open(WORD_SET2, ">word_set2.txt") || die "Can't open word_set2.txt" ;
while (@word_set2)
{  $_ = shift(@word_set2);
   print WORD_SET2 "$_\n";
}
close(WORD_SET2);
exit 0;