diff --git a/bin/jekyll-planet.rb b/bin/jekyll-planet.rb
index 14687c4..4005fdc 100644
--- a/bin/jekyll-planet.rb
+++ b/bin/jekyll-planet.rb
@@ -1,101 +1,107 @@
 # encoding: utf-8
 require 'pluto/models'
 require 'nokogiri'
 require "stringex"
 
 puts "db settings:"
 @db_config = {
   adapter: 'sqlite3',
   database: './planet.db'
 }
 
 pp @db_config
 
 
 def run( args )
   unless File.exists?( @db_config[:database])
     puts "** error: database #{@db_config[:database]} missing; please check pluto documention for importing feeds etc."
     exit 1;
   end
 
   Pluto.connect( @db_config )
 
-  Pluto::Model::Item.latest.where.not(title: '', published: nil).limit(1000).each_with_index do |item,i|
+  Pluto::Model::Item.latest.where.not(title: '').limit(1000).each_with_index do |item,i|
+    if item.published == nil and item.updated == nil
+      next
+    end
+    if item.published == nil and item.updated != nil
+      item.published = item.updated
+    end
     puts "[#{i+1}] #{item.title}"
 
     generate_blog_post( item )
   end
 end
 
 def generate_frontmatter( data )
 
   frontmatter = ''
   data.each do |key, value|
     spaces = ' ' * (data.keys.map(&:length).max + 1 - key.length) unless value.is_a?(Array)
     output = value
     output = "#{value.dump}" if value.is_a?(String)
     output = "\n  - \"#{value.join("\"\n  - \"")}\"" if value.is_a?(Array)
     frontmatter += "#{key}:#{spaces}#{output}\n"
   end
   frontmatter
 
 end
 
 def generate_blog_post( item )
 
   posts_root = "./_posts"
 
   FileUtils.mkdir_p( posts_root )  ## make sure path exists
 
   ## Note:
   ## Jekyll pattern for blogs must follow
   ##  2014-12-21-  e.g. must include trailing dash (-)
   fn = "#{posts_root}/#{item.published.strftime('%Y-%m-%d')}-#{item.title.to_url}.html"
   # Check for author tags
 
   data = {}
   data["title"] = item.title.gsub('"','\"') unless item.title.empty?
   data["created_at"] = item.published if item.published
   data["updated_at"] = item.updated if item.updated
   data["guid"] = item.guid unless item.guid.empty?
   data["author"] = item.feed.title unless item.feed.title.empty?
   data["avatar"] = item.feed.avatar if item.feed.avatar
   data["link"] = item.feed.link unless item.feed.link.empty?
   data["rss"] = item.feed.feed unless item.feed.feed.empty?
   data["tags"] = [ item.feed.location ? item.feed.location : "en" ]
   data["original_link"] = item.url if item.url
   item.feed.author.split.each do |contact|
     if contact.include?(':')
       part = contact.split(':')
       data[part.shift] = part.join(':')
     else
       data[contact] = true
     end
   end if item.feed.author
   frontmatter = generate_frontmatter(data)
 
   File.open( fn, 'w' ) do |f|
     f.write "---\n"
     f.write frontmatter
     f.write "---\n"
 
     # There were a few issues of incomplete html documents, nokogiri fixes that
     html = ""
     if item.content
       html = Nokogiri::HTML::DocumentFragment.parse(item.content).to_html
     elsif item.summary
       html = Nokogiri::HTML::DocumentFragment.parse(item.summary).to_html
     else
       ## warn: not content found for feed
     end
     # Liquid complains about curly braces
     html.gsub!("{", "{")
     html.gsub!("{", "}")
     html.gsub!(/(?<=src=[\"\'])\/(?!\/)/, "#{/\/\/.*?(?=\/|$)/.match(item.feed.link)[0]}/")
     html.gsub!(/(?<=src=[\"\'])https?:/, "")
     f.write html
   end
 end
 
 run( ARGV )