#!/usr/bin/perl

# Script to build nugget pyramids out of individual vital/okay
# judgments.  Please refer to the following citation:
#
#   Jimmy Lin and Dina Demner-Fushman. Will Pyramids Built of Nuggets
#   Topple Over? Proceedings of the 2006 Human Language Technology
#   Conference and the North American Chapter of the Association for
#   Computational Linguistics Annual Meeting (HLT/NAACL 2006), June
#   2006, New York City, New York.
#

@judgments = ( "trec2003.nuggets.0", 
	       "trec2003.nuggets.1", 
	       "trec2003.nuggets.2", 
	       "trec2003.nuggets.3", 
	       "trec2003.nuggets.4", 
	       "trec2003.nuggets.5", 
	       "trec2003.nuggets.6", 
	       "trec2003.nuggets.7", 
	       "trec2003.nuggets.8", 
	       "trec2003.nuggets.9", 
);

# first, load all nuggets
foreach $curfile ( @judgments ) {
    #print "loading $curfile...\n";
    open(FILE, "< ./$curfile");
    while ( <FILE> ) {
	if ( /([\d.]+)\s+(\d+)\s+(vital|okay)/ ) {
	    $jud = $3;
	    $qid = $1;
	    $nid = $2;

	    #print "$qid#$nid\t$jud\n";
	    $raw{$curfile}{"$qid#$nid"} = $jud;
	}
    }
}

# iterate to count vitals
open(FILE, "< $judgments[0]");
while ( <FILE> ) {
    
    if ( /([\d.]+)\s+(\d+)\s+(vital|okay)\s+([^\n]+)/ ) {
	$qid = $1;
	$nid = $2;
	$nugget = $4;

	$cnt = 0;
	foreach $f ( @judgments ) {
	    $jud = $raw{$f}{"$qid#$nid"};
	    #print "$f $jud\n";
	    $cnt++ if ( $jud eq 'vital' );
	}

	$vital_cnt{$qid}{$nid} = $cnt;
	#print "$qid\t$nid\t$cnt\t$nugget\n";
    }
}
close(FILE);


# finally, normalize and print out
open(FILE, "< $judgments[0]");
while ( <FILE> ) {
    
    if ( /([\d.]+)\s+(\d+)\s+(vital|okay)\s+([^\n]+)/ ) {
	$qid = $1;
	$nid = $2;
	$nugget = $4;

	$max = 0;
	foreach $k ( keys %{$vital_cnt{$qid}} ) {
	    $max = $vital_cnt{$qid}{$k} if $vital_cnt{$qid}{$k} > $max;
	}
	$score = $vital_cnt{$qid}{$nid} / $max;
	print "$qid\t$nid\t$score\t$nugget\n";

    } else {
	print;
    }
}
close(FILE);