#!/usr/bin/perl

if (!exists($ARGV[0])) { print "USAGE: ./parsebqcsvtogeogephi.pl INFILE\n"; exit; }

###################################################################
#load our geocrossref mapping...
open(FILE, "./COUNTRY-GEO-LOOKUP.TXT");
while(<FILE>) {
    ($name, $lat, $long, $cc) = split/\t/, $_; $cc=~s/\s+$//;
    if (abs($lat) > 0 || abs($long) > 0) {
	$CCNAME{$cc} = $name;
	$CCATT{$cc} = "<attvalues><attvalue for=\"country\" value=\"$name\"/><attvalue for=\"latitude\" value=\"$lat\"/><attvalue for=\"longitude\" value=\"$long\"/></attvalues>";
    }
}
close(FILE);
###################################################################

###################################################################
#print our GEFX opening...
open(OUT, ">$ARGV[0].gexf");
print OUT "
<gexf xmlns=\"http://www.gexf.net/1.2draft\" version=\"1.2\" xmlns:viz=\"http://www.gexf.net/1.2draft/viz\" xmlns:xsi=\"http://www.w3.org/2001/XMLSchema-instance\" xsi:schemaLocation=\"http://www.gexf.net/1.2draft http://www.gexf.net/1.2draft/gexf.xsd\">  
<graph mode=\"static\" defaultedgetype=\"undirected\"><attributes class=\"node\" mode=\"static\">
<attribute id=\"city\" title=\"City\" type=\"string\"/>
<attribute id=\"latitude\" title=\"latitude\" type=\"double\"/>
<attribute id=\"longitude\" title=\"longitude\" type=\"double\"/>
</attributes>
<nodes>
";
###################################################################

###################################################################
#load the input CSV and write the nodes and hash up the edges...
$MASTERTOTAL = 0;
open(FILE, $ARGV[0]);
$NODEID = 0;
while(<FILE>) {
    ($from, $to, $count) = split/,/, $_; $count=~s/\s+$//;
    if ($count > 0 && exists($CCATT{$from}) && exists($CCATT{$to})) {
    	if (!exists($NODEIDS{$from})) { $NODEIDS{$from} = $NODEID; $NODECNT{$from}+=$count; $NODEID++; }
    	if (!exists($NODEIDS{$to})) { $NODEIDS{$to} = $NODEID; $NODECNT{$to}+=$count; $NODEID++; }
	$EDGES{"$NODEIDS{$from}\t$NODEIDS{$to}"} = $count;
	$MASTERTOTAL += $count;
    }
}
close(FILE);

#and write the nodes to disk...
foreach $node (sort keys %NODEIDS) {
    print OUT "\t<node id=\"$NODEIDS{$node}\" label=\"$CCNAME{$node}\" weight=\"$NODECNT{$node}\">$CCATT{$node}</node>\n";
}
###################################################################
print OUT "</nodes>\n<edges>\n";
#now write the edges now that we've assigned all of the nodes their nodeids...
$EDGEID = 0;
foreach $edge (sort keys %EDGES) {
    ($node1, $node2) = split/\t/, $edge;
    $weight = $EDGES{$edge} / $MASTERTOTAL;
    print OUT "<edge id=\"$EDGEID\" source=\"$node1\" target=\"$node2\" weight=\"$weight\"></edge>\n";
    $EDGEID++;
}
###################################################################

###################################################################
#print our GEFX closing...
print OUT "
</edges>
</graph>
</gexf>
";
close(OUT);
###################################################################

#done!