I broke a long one-liner in matching.html into it's own perl program …

…(dist.pl)

and changed it in minor ways to get it working.

dist.pl

@@ -3,7 +3,9 @@
 # This program is listed as a one-liner in matching.html. One-liners are hard to debug,
 # though, so I'm pulling it out into a normal program.  --jhannah 20100311
 
-$dist = 1000;
+use strict;
+
+my (@data, @olddata, $dist);
 while (<>) {
    @data = split; 
    if ($data[3] eq "-") {
@@ -13,13 +15,13 @@
          }
          print "@data[0..5] 1000 $data[6]\n"
       } else {
-         $dist = $data[2] - $olddata[2]; 
          print "@olddata[0..5] $dist $data[6]\n@data[0..5] $dist $data[6]\n";
       } 
+   } else {
+      $dist = @olddata ? $data[2] - $olddata[2] : 1000;
    }
    @olddata = @data;
 }
-print "@olddata[0..5] 1000 $data[6]\n";ii
-
+print "@olddata[0..5] 1000 $data[6]\n";
 
 

matching.html

@@ -79,9 +79,9 @@ <h2>
 <p>
 Because we'll want to exclude analysis of sites too close to each other (and so they would interfere with each other) we want to add a column to the data indicating distance to the neighboring site. We did this with command line perl code like the following (note that this code assumes that the upstream sites precede the downstream sites in the input file):
 </p>
-<p>
-cat CCGG_tags_18 | perl -ne 'BEGIN {$dist = 1000;} @data = split; if ($data[3] eq "-") {if ($data[1] ne $olddata[1]) { if (@olddata) { print "@olddata[0..5] 1000 $data[6]\n";}; print "@data[0..5] 1000 $data[6]\n";} else {$dist = $data[2] - $olddata[2]; print "@olddata[0..5] $dist $data[6]\n@data[0..5] $dist $data[6]\n";} }; @olddata = @data; END {print "@olddata[0..5] 1000 $data[6]\n";}' &gt CCGG_tags_18_withdist
-</p>
+<pre>
+perl dist.pl CCGG_tags_18 &gt CCGG_tags_18_withdist
+</pre>
 <p>
 This ends up looking like this:
 </p>