patch for bogotune

[m at mo.optusnet.com.au]

My corpus is held in MH style folders, but owinging to the number of
messages, the directories are 'hashed' to keep the size on individual
directories manageable.

bogotune hates this. :)

Attached is a patch that expands the MH support to handle nested
folders. Should be fully backward compatable.


Michael.


--- /usr/share/bogofilter/tuning/bogotune	Fri Jun 27 05:31:34 2003
+++ ./bogotune	Sun Jul 20 13:48:09 2003
@@ -138,11 +138,11 @@
 if(-d $spfiles[0]) {
     $msgformat = "MH"; $scount = 0;
     foreach my $dir (@spfiles) {
-	$scount += `ls $dir/[0-9]* 2>/dev/null | wc -l`;
+	$scount += `find $dir -type f -name '[0-9]*' 2>/dev/null | wc -l`;
     }
     $ncount = 0;
     foreach my $dir (@nsfiles) {
-	$ncount += `ls $dir/[0-9]* 2>/dev/null | wc -l`;
+	$ncount += `find $dir -type f -name '[0-9]*' 2>/dev/null | wc -l`;
     }
 } else {
     my $cmd = join(" ", "cat", @spfiles, "| grep -c '^From ' |");
@@ -178,28 +178,24 @@
     } else {
 	unlink($spwork);
 	foreach my $dir (@spfiles) {
-	    opendir(DH, $dir) or yuk(7, "Problem processing spam files");
-	    my @msgs = readdir(DH); closedir(DH);
-	    foreach my $msg(@msgs) {
-		if($msg =~ /^[0-9]/) {
-		    my $cmd = join(" ", "cat $dir/$msg | bogol", $bogodir,
+	    open FL, "find $dir -type f -name '[0-9]*'|" or yuk(7, "Problem processing spam files");
+	    foreach my $msg (<FL>) {
+		chomp($msg);
+		my $cmd = join(" ", "cat $msg | bogol", $bogodir,
 			$cf, ">>$spwork");
-		    system($cmd) == 0 or
+		system($cmd) == 0 or
 			yuk(7, "Problem writing spam msg-count file");
-		}
 	    }
 	}
 	unlink($nswork);
 	foreach my $dir (@nsfiles) {
-	    opendir(DH, $dir) or yuk(7, "Problem processing nonspam files");
-	    my @msgs = readdir(DH); closedir(DH);
-	    foreach my $msg (@msgs) {
-		if($msg =~ /^[0-9]/) {
-		    my $cmd = join(" ", "cat $dir/$msg | bogol", $bogodir,
+	    open FL, "find $dir -type f -name '[0-9]*'|" or yuk(7, "Problem processing nonspam files");
+	    foreach my $msg (<FL>) {
+		chomp($msg);
+		my $cmd = join(" ", "cat $msg | bogol", $bogodir,
 			$cf, ">>$nswork");
-		    system($cmd) == 0 or
+		system($cmd) == 0 or
 			yuk(7, "Problem writing nonspam msg-count file");
-		}
 	    }
 	}
     }