diff --git a/examples/schemaless-mysql/mysql_interceptor.rb b/examples/schemaless-mysql/mysql_interceptor.rb
index 184470d..d7ea1e5 100644
--- a/examples/schemaless-mysql/mysql_interceptor.rb
+++ b/examples/schemaless-mysql/mysql_interceptor.rb
@@ -1,11 +1,13 @@
 require "lib/em-proxy"
 require "em-mysql"
+require "stringio"
 require "fiber"
 
 Proxy.start(:host => "0.0.0.0", :port => 3307) do |conn|
   conn.server :mysql, :host => "127.0.0.1", :port => 3306, :relay_server => true
 
   QUERY_CMD = 3
+  MAX_PACKET_LENGTH = 2**24-1
 
   # open a direct connection to MySQL for the schema-free coordination logic
   @mysql = EventMachine::MySQL.new(:host => 'localhost', :database => 'noschema')
@@ -25,9 +27,9 @@
 
         case query.first
         when "create" then
-          # allow schemaless table creation, ex: 'create table posts'
-          # by creating a table with a single id for key storage, aka
-          # rewrite to: 'create table posts (id varchar(255))'. all
+          # Allow schemaless table creation, ex: 'create table posts'
+          # By creating a table with a single id for key storage, aka
+          # rewrite to: 'create table posts (id varchar(255))'. All
           # future attribute tables will be created on demand at
           # insert time of a new record
           overload = "(id varchar(255), UNIQUE(id));"
@@ -37,7 +39,7 @@
           p [:create_new_schema_free_table, query, data]
 
         when "insert" then
-          # overload the INSERT syntax to allow for nested parameters
+          # Overload the INSERT syntax to allow for nested parameters
           # inside the statement. ex:
           #   INSERT INTO posts VALUE("post_id_1", (
           #     ("author", "Ilya Grigorik"),
@@ -74,7 +76,7 @@
               q = @mysql.query(attr_sql)
               q.errback  { |res|
                 # if the attribute table for this model does not yet exist then create it!
-                # - yes, there is a race condition here, add a fiber later
+                # - yes, there is a race condition here, add fiber logic later
                 if res.is_a?(Mysql::Error) and res.message =~ /Table.*doesn\'t exist/
 
                   table_sql = "create table #{table}_#{value[1]} (id varchar(255), value varchar(255), UNIQUE(id))"
@@ -96,32 +98,65 @@
           end
 
         when "select" then
-          attrs = sql.match(/select(.*?)from/)[1].strip.split(',')
-          p [:select, attrs]
-
-          tables = @mysql.query("show tables like 'posts_%'")
-          tables.callback {|res| fiber.resume(res.all_hashes.collect(&:values).flatten) }
-          tables = Fiber.yield + [table]
-          
-          p [:select_tables, tables]
-#          query = tables
-
           #  select posts.id as id, posts_author.value as author FROM posts
           #   LEFT OUTER JOIN posts_author ON posts_author.id = posts.id
           #   WHERE posts.id = "ilya";
           # select posts.id as id, posts_author.value as author FROM posts LEFT OUTER JOIN posts_author ON posts_author.id = posts.id WHERE posts.id = "ilya";
 
+          select = sql.match(/select(.*?)from\s([^\s]+)/)
+          where  = sql.match(/where\s([^=]+)\s?=\s?'?"?([^\s'"]+)'?"?/)
+          attrs, table = select[1].strip.split(','), select[2]
+          key = where[2] if where
+
+          p [:select, select, attrs, where]
+
+          tables = @mysql.query("show tables like '#{table}_%'")
+          tables.callback { |res|
+            fiber.resume(res.all_hashes.collect(&:values).flatten.collect{ |c|
+                c.split('_').last 
+              })
+          }
+          tables = Fiber.yield
+          
+          p [:select_tables, tables]
+
+          # build the select statements, hide the tables behind each attribute
+          join =  "select #{table}.id as id "
+          tables.each do |column|
+            join += " , #{table}_#{column}.value as #{column} "
+          end
+
+          # add the joins to stich it all together
+          join += " FROM #{table} "
+          tables.each do |column|
+            join += " LEFT OUTER JOIN #{table}_#{column} ON #{table}_#{column}.id = #{table}.id "
+          end
+
+          join += " WHERE #{table}.id = '#{key}' " if key
+
+          query = [join]
+          overhead = join.size + 1
 
+          p [:join_query, join]
         end
 
         # repack the query data and forward to server
-        data = [overhead, chunks, seq].pack("CvC") + [type, query.join(" ")].pack("Ca*")
+        # - have to split message on packet boundaries
+ 
+        seq, data = 0, []
+        query = StringIO.new([type, query.join(" ")].pack("Ca*"))
+        while q = query.read(MAX_PACKET_LENGTH)
+          data.push [q.length % 256, q.length / 256, seq].pack("CvC") + q
+          seq = (seq + 1) % 256
+        end
 
-        p [:final_query, data]
+        p [:final_query, data, chunks, overhead]
         puts "-" * 100
       end
 
-      conn.relay_to_servers(data)
+      [data].flatten.each do |chunk|
+        conn.relay_to_servers(chunk)
+      end
 
       :async # we will render results later
     }
@@ -130,4 +165,8 @@
   end
 end
 
-# create table #{table}_#{value[1]} (id varchar(255), value varchar(255), UNIQUE(id))
\ No newline at end of file
+#
+# INSERT INTO posts VALUE("igvita2", (('org', 'PostRank'),('nickname', 'igrigorik'),('title', 'hello world')));
+# select * from posts;
+# select * from posts where id = 'igvita';
+#
\ No newline at end of file