#!/usr/bin/perl

#@grammars = ("/afs/cs/project/avenue-1/Avenue/Transfer/Chinese/learned/selectedrules.trf" , "/afs/cs/project/avenue-1/Avenue/Transfer/Chinese/learned/selectedlex.trf");
@grammars = ("/tmp/vamshi-rules-3-gb.trf");
foreach $grammar (@grammars) {
    open(GRA, $grammar) or die $!;
    $ruletext = "";
    while ($line = <GRA>) {
	if ($line =~ m/\{(\w+,\s*\d+)\s*\}/) {
	    $grammarids{uc($1)} = 1;
	    $grammarscores{uc($1)} = $score;
	} elsif ($line =~ m/;;Score::(\d+)/) {
	    $score = $1;
	}
    }
    close(GRA);
}

$hypfile = "/shared/data/Chinese/mt03_chinese_evlset_v0-xfer3-071117-1.txt";
#$hypfile = "/shared/data/Chinese/mt03_chinese_evlset_v0-xfer3-070718-1.txt";

open(HYP, $hypfile) or die $!;
while ($line = <HYP>) {
    if ($line =~ m/^\d+ (\d+)\t/) {
	if ($1 == 0) {
	    $record = 1;
	} else {
	    $record = 0;
	}
    }

    (@ids) = ($line =~ m/\((\w+,\d+) /g);
    foreach $id (@ids) {
	$idcount{uc($id)}++ if $record;
    }
}

foreach $id (sort {$idcount{$b} <=> $idcount{$a}} keys %idcount) {
    if (defined($grammarids{$id})) {
	print $id, "\t", $idcount{$id}, "\t", $grammarscores{$id}, "\n";
    }
}