Added comments

lib/tasks/tag_clustering.rake

@@ -22,12 +22,14 @@ namespace :tag do
   task :generate_clusters => [:environment] do
     # DEBUG = true
     graph_path = TagSearchService.file_path+".net"
+    # Create hash of tag co occurrence
     hash = create_hash()
-
+    # Create a graph infomap can read
     create_pajek_net_graph(hash, graph_path)
 
     infomap_ftree(graph_path, TagSearchService.root_dir)
 
+    # Cluster needs to be read from disk again, so clear cache of TagSearchService
     Rails.cache.delete(TagSearchService::CACHE_KEY)
   end # task
 
@@ -42,28 +44,25 @@ namespace :tag do
       # for each lo, count tags and tag pairs and add to hash
       # if id1 <= id2
       lo.tags.each.with_index do |t, i|
-        # initialize value
+        # initialize values
         hash[t.id] = {} if hash[t.id].nil?
         hash[t.id][t.id] = 0 if hash[t.id][t.id].nil?
 
-        hash[t.id][t.id] += 1
+        hash[t.id][t.id] += 1 # add self occurrence
 
-        # for each next tags (with higher index)
+        # for each next tags (with higher index than t)
         lo.tags.drop(i+1).each do |t2|
-          # [t1][t2], t1 should always be lower
+          # swap if necessary ([t1][t2], t1 should always have lower id)
           if t.id > t2.id
-            # swaps
+            # swaps t with t2
             t, t2 = t2, t
             # check nil
             hash[t.id] = {} if hash[t.id].nil?
           end
-
           # initialize value
-          if hash[t.id][t2.id].nil?
-            hash[t.id][t2.id] = 0
-          end
+          hash[t.id][t2.id] = 0 if hash[t.id][t2.id].nil?
 
-          hash[t.id][t2.id] += 1
+          hash[t.id][t2.id] += 1 # add co occurrence
         end
       end
     end