wordpress.rb

require 'rubygems'
require 'sequel'
require 'fileutils'
require 'yaml'

# 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("_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, \
                      ID, \
                      guid \
               FROM #{table_prefix}posts \
               WHERE post_status = 'publish' AND \
                     post_type = 'post'"

      # Fetch all tags for a given POST ID
      tags_query = "select tm.term_id,tm.name from wp_term_relationships tr
                    inner join wp_term_taxonomy tt on tr.term_taxonomy_id = tt.term_taxonomy_id
                    inner join wp_terms tm on tm.term_id=tt.term_id
                    where tr.object_id=%d and tt.taxonomy = 'post_tag'";

      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]
        content = post[:post_content]
        name = "%02d-%02d-%02d-%s.markdown" % [date.year, date.month, date.day,
                                               slug]
        tags = []
        db[tags_query % post[:ID]].each do |tag|
            tags << tag[:name].to_s.gsub('+','').downcase
        end

        # Process content to rewrite some URLs
        baseurl = "http://www.manu-j.com/blog/wp-content/uploads/"
        content.gsub!(baseurl,"http://blog.manu-j.com/uploads/")

        baseurl = "http://static.manu-j.com/blog/wp-content/uploads/"
        content.gsub!(baseurl,"http://blog.manu-j.com/uploads/")

        content.scan(/\[TABLE=(\d+)\]/i).each do |table_id|
          table_id = table_id.first
          table_query = "SELECT row_id, value FROM wp_golfresult
            WHERE table_id =  #{table_id} ORDER BY row_id"
          table = "<table><thead><tr>\n"
          previous_row_id = -1
          total_column_count = 0
          current_column = 0
          aligns =  {}
          widths =  {}
          align_expand = {'L' => 'left', 'C' => 'center', 'R' => 'right'}
          db[table_query].each do |row|
            row_id = row[:row_id].to_i

            if previous_row_id != row_id
              current_column = 0
              if previous_row_id == 1
                table += "</tr></thead><tbody><tr>\n"
              elsif previous_row_id > 1
                table += "</tr><tr>\n"
              end
            end
            current_column += 1

            case row_id
            when -1
              aligns[current_column] = align_expand[row[:value]]
            when 0
              widths[current_column] = row[:value]
            when 1
              table += "  <th align='#{aligns[current_column]}' width='#{widths[current_column]}px'>#{row[:value]}</th>\n"
            else
              table += "  <td align='#{aligns[current_column]}' width='#{widths[current_column]}px'>#{row[:value]}</td>\n"
            end

            previous_row_id = row_id

          end
          table += '</tr></tbody></table>'
          content.gsub!(/\[TABLE=#{table_id}\]/i,table)
        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],
           'permalink' => "/#{slug.downcase.gsub(/\s/,'-')}/#{post[:ID]}/",
           'tags' => tags,
           'date' => date
         }.delete_if { |k,v| v.nil? || v == '' }.to_yaml

        # Write out the data and content to file
        File.open("_posts/#{name}", "w") do |f|
          f.puts data
          f.puts "---"
          f.puts content
        end
      end

    end
  end
end