add a literal dupe check for the comments in spots

[spider.git] / perl / Spot.pm

diff --git a/perl/Spot.pm b/perl/Spot.pm
index 61bc1a25959424bce67ced5ee6a37660e949ccb7..f331d105b540ae1cd2e4b414c631ffcfbf97eca4 100644 (file)
--- a/perl/Spot.pm
+++ b/perl/Spot.pm
@@ -36,7 +36,7 @@ $defaultspots = 10;                           # normal number of spots to return
 $maxdays = 100;                                # normal maximum no of days to go back
 $dirprefix = "spots";
 $duplth = 20;                                  # the length of text to use in the deduping
-$dupage = 3*3600;               # the length of time to hold spot dups
+$dupage = 1*3600;               # the length of time to hold spot dups
 $maxcalllth = 12;                               # the max length of call to take into account for dupes
 $filterdef = bless ([
                          # tag, sort, field, priv, special parser 
@@ -319,7 +319,7 @@ sub readfile($)
 # enter the spot for dup checking and return true if it is already a dup
 sub dup
 {
-       my ($freq, $call, $d, $text, $by) = @_; 
+       my ($freq, $call, $d, $text, $by, $cty) = @_; 
 
        # dump if too old
        return 2 if $d < $main::systime - $dupage;
@@ -328,23 +328,38 @@ sub dup
        $d = int ($d / 60);
        $d *= 60;
 
+       # remove SSID or area
+       $by =~ s|[-/]\d+$||;
+       
        $freq = sprintf "%.1f", $freq;       # normalise frequency
        $call = substr($call, 0, $maxcalllth) if length $call > $maxcalllth;
 
        chomp $text;
        $text =~ s/\%([0-9A-F][0-9A-F])/chr(hex($1))/eg;
-       $text = unpad($text);
-       $text = substr($text, 0, $duplth) if length $text > $duplth; 
+       $text = uc unpad($text);
+       if ($cty && $text && length $text <= 4) {
+               unless ($text =~ /^C?Q/ || $text =~ /^\d+$/) {
+                       my @try = Prefix::cty_data($text);
+                       $text = "" if $cty == $try[0];
+               }
+       }
+       my $otext = $text;
        $text = pack("C*", map {$_ & 127} unpack("C*", $text));
-       $text =~ s/[^a-zA-Z0-9]//g;
-       my $ldupkey = "X$freq|$call|" . uc $text;
-       my $sdupkey = "X$freq|$call|$by";
+       $text =~ s/[^\w]//g;
+       $text = substr($text, 0, $duplth) if length $text > $duplth; 
+       my $ldupkey = "X$freq|$call|$by|$text";
        my $t = DXDupe::find($ldupkey);
-       return 1 if $t && $t - $main::systime > 0;      
-    $t = DXDupe::find($sdupkey);
-       return 1 if $t && $t - $main::systime > 0;      
+       return 1 if $t && $t - $main::systime > 0;
+       DXDupe::add($ldupkey, $main::systime+$dupage);
+       $otext = substr($otext, 0, $duplth) if length $otext > $duplth; 
+       $ldupkey = "X$freq|$call|$by|$otext";
+       $t = DXDupe::find($ldupkey);
+       return 1 if $t && $t - $main::systime > 0;
        DXDupe::add($ldupkey, $main::systime+$dupage);
-       DXDupe::add($sdupkey, $main::systime+$dupage);
+#      my $sdupkey = "X$freq|$call|$by";
+#      $t = DXDupe::find($sdupkey);
+#      return 1 if $t && $t - $main::systime > 0;      
+#      DXDupe::add($sdupkey, $main::systime+$dupage);
        return 0;
 }