pimped octopress wordpress migrator script

wordpress.rb

require 'rubygems'
require 'sequel'
require 'fileutils'
require 'yaml'
require 'active_support/inflector'

# pimped by http://github.com/benben
# original script from here: https://github.com/mojombo/jekyll/blob/master/lib/jekyll/migrators/wordpress.rb
# some parts stolen from here: http://vitobotta.com/how-to-migrate-from-wordpress-to-jekyll/#importing-from-wordpress

# NOTE: This converter requires Sequel and the MySQL gems.
# The MySQL gem can be difficult to install on OS X. Once you have MySQL
# installed, running the following commands should work:
# $ sudo gem install sequel
# $ sudo gem install mysql -- --with-mysql-config=/usr/local/mysql/bin/mysql_config

module Jekyll
  module WordPress
    def self.process(dbname, user, pass, host = 'localhost', table_prefix = 'wp_')
      db = Sequel.mysql(dbname, :user => user, :password => pass, :host => host, :encoding => 'utf8')

      FileUtils.mkdir_p("source/_posts")

      # Reads a MySQL database via Sequel and creates a post file for each
      # post in wp_posts that has post_status = 'publish'. This restriction is
      # made because 'draft' posts are not guaranteed to have valid dates.
      query = "SELECT post_title, \
                      post_name, \
                      post_date, \
                      post_content, \
                      post_excerpt, \
                      post_author, \
                      post_status, \
                      ID, \
                      guid \
               FROM #{table_prefix}posts \
               WHERE post_status = 'publish' AND \
                     post_type = 'post'"

      categories_and_tags_query = <<-EOS

        SELECT         t.taxonomy, term.name, term.slug
        FROM             #{table_prefix}term_relationships AS tr
        INNER JOIN   #{table_prefix}term_taxonomy AS t ON t.term_taxonomy_id = tr.term_taxonomy_id
        INNER JOIN   #{table_prefix}terms AS term ON term.term_id = t.term_id
        WHERE            tr.object_id = %d
        ORDER BY   tr.term_order

      EOS

      db[query].each do |post|
        # Get required fields and construct Jekyll compatible name.
        title = post[:post_title]
        slug = post[:post_name]
        date = post[:post_date]
        status = post[:post_status]
        author = db["SELECT display_name FROM #{table_prefix}users WHERE ID = #{post[:post_author]} LIMIT 1"][:display_name][:display_name]
        content = post[:post_content]

        content.gsub!(/<pre lang="(.+)">(\n|$|\s)/, '``` \1')
        content.gsub!(/<pre lang="(.+)">/, '``` \1' + "\n")
        content.gsub!(/(\n|$|\s)<\/pre>/, '\1```')
        content.gsub!(/<\/pre>/, "\n" + '```')

        name = "%02d-%02d-%02d-%s.markdown" % [date.year, date.month, date.day, slug]
        categories = []
        post_tags  = []

        db[categories_and_tags_query % post[:ID]].each do |category_or_tag|
          eval(category_or_tag[:taxonomy].pluralize) << category_or_tag[:name]
        end

        # Get the relevant fields as a hash, delete empty fields and convert
        # to YAML for the header.
        data = {
           'layout' => 'post',
           'title' => title.to_s,
           'excerpt' => post[:post_excerpt].to_s,
           'wordpress_id' => post[:ID],
           'wordpress_url' => post[:guid],
           'date' => date,
           'comments' => true,
           'author' => author,
           'categories' => categories,
           'tags' => post_tags
         }.delete_if { |k,v| v.nil? || v == '' }.to_yaml

        # Write out the data and content to file
        File.open("source/#{status == 'publish' ? '_posts' : '_drafts'}/#{name}", "w") do |f|
          f.puts data
          f.puts "---"
          f.puts content
        end
      end
    end
  end
end