wp-xml-import.rb

# -*- encoding: utf-8 -*-
require 'fileutils'
require 'date'
require 'yaml'
require 'rexml/document'
include REXML

doc = Document.new File.new(ARGV[0])

FileUtils.mkdir_p "_posts"

doc.elements.each("rss/channel/item[wp:status = 'publish' and wp:post_type = 'post']") do |e|
    post = e.elements
    slug = post['wp:post_name'].text
    date = DateTime.parse(post['wp:post_date'].text)
    name = "%02d-%02d-%02d-%s.markdown" % [date.year, date.month, date.day, slug]
    date_string = "#{date.year}-#{date.month}-#{date.day} #{date.hour}:#{date.min}"

    # content
    content = post['content:encoded'].text.encode("UTF-8")
    content_array = []
    bq = 0
    content.split(/\n/).each do |t|
      # code
      t = t.gsub(/(\[|<)(php|code)(\]|>)(.*)(\[|<)\/(php|code)(\]|>)/, "\n```\n" + '\4' + "\n```\n")

      t = t.gsub(/(\[|<)(php|code)(\]|>)(.*)$/, "\n```\n" + '\4')

      t = t.gsub(/\[(php|code) ([^=]*)=\"([^\"]*)\"\](.*)\[\/(php|code)\]/, "\n```\n" + '\4'+"\n```")
      t = t.gsub(/\[(php|code) ([^=]*)=\"([^\"]*)\"\](.*)$/, "\n```\n" + '\4')
      t = t.gsub(/^(.*)(\[|<)\/(php|code)(\]|>)/, '\1'+"\n```")
      t = t.gsub(/^(\s*)<li>([^<]*)<\/li>/, '* \2')

      if t =~ /^<\/?ul>$/
        next
      end
      if t =~ /^<\/?ol>$/
        next
      end
      content_array << t
    end
    content = content_array.flatten.join("  \n")

    content = content.gsub(/<p>(.*)<\/p>/, '\1')
    content = content.gsub(/&lt;/, '<')
    content = content.gsub(/&gt;/, '>')
    content = content.gsub(/&quot;/, '"')

    content = content.gsub(/<br \/>/, '  '+"\n")
    content = content.gsub(/<strong>([^<]*)<\/strong>/, '**\1**')
    content = content.gsub(/<b>([^<]*)<\/b>/, '**\1**')
    content = content.gsub(/<a href=[\"\']([^\"\']*)[\"\']>([^<]*)<\/a>/, '[\2](\1)')

    (1..5).each do |i|
        content = content.gsub(/<h#{i}>([^<]*)<\/h#{i}>/, ('#'*i) + ' \1')
        content = content.gsub(/<h#{i} class=([^>]*)>([^<]*)<\/h#{i}>/, ('#'*i) + ' \2')
    end

    # categories & tags
    category_string = ""
    tags_array = []
    post.each("category") do |e|
      if e.attributes["domain"] == "category"
        category_string = e.attributes["nicename"].to_s
      end
      if e.attributes["domain"] == "post_tag"
        tags_array << e.attributes["nicename"].to_s
      end
    end

    title_text = post['title'].text

    puts "Converting: #{name}"

    data = {
        'layout' => 'post',
        'title' => title_text,
        'date' => date_string,
        'comments' => false,
        'categories' => category_string,
        'tags' => tags_array,
     }.delete_if { |k,v| v.nil? || v == ''}.to_yaml

     File.open("_posts/#{name}", "w") do |f|
         f.puts data
         f.puts "---"
         f.puts content
     end
 
end