Updated documentation, the content is now converted from html to markdown, fixed problems with files with some characters in their names and in their categories and fixed categories with spaces.

blogspot_to_jekyll.rb

#!/usr/bin/env ruby
#
# Convert blogger (blogspot) posts to jekyll posts
#
# Basic Usage
# -----------
#   In Windows, go to the folder where this file is and execute:
#   ruby blogspot_to_jekyll.rb feed_url
#
#   In Linux, go to the folder where this file is and execute:
#   ./blogspot_to_jekyll.rb feed_url
#
#  where `feed_url` can have the following format:
#  http://{your_blog_name}.blogspot.com/feeds/posts/default
#
#  The new posts will be created in "where_this_file_is/_posts"
#
# Requirements
# ------------
#
#  * feedjira: https://github.com/feedjira/feedjira
#  * reverse_markdown: https://github.com/xijo/reverse_markdown
#
# Notes
# -----
#
#  * Make sure Blogger shows full output of article in feeds.
#  * Make sure the feed shows all the posts you want to import. If not, add "?max-results=500" to your feed url.
#  * If you have more than 500 post to import, check this:
#    http://too-clever-by-half.blogspot.kr/2011/12/blog-feed-500-post-limit-for-more-than.html
#  * Commenting on migrated articles will be set to false by default.
#  * If your title or your categories have some characters like "á", "é", "í",... you have to replace them manually
#    or your page will have problems to load.
#  * The following characters will be deleted from the title when importing: / : " , and ..
#    In case of categories, only the colon will be deleted.

require 'rubygems' #if RbConfig['host_os'].start_with? "darwin"
require 'feedjira'
require 'date'
require 'optparse'
require 'reverse_markdown'

def parse_post_entries(feed, verbose)
  posts = []
  feed.entries.each do |post|
    obj = Hash.new
    created_datetime = post.published
    creation_date = Date.parse(created_datetime.to_s)
    title = post.title.delete('\/|\:|\"|\,|\.')
    file_name = creation_date.to_s + "-" + title.split(/  */).join("-").delete('\/|\:|\"|\,|\.') + ".md"
    content = ReverseMarkdown.convert(post.content, unknown_tags: :drop, github_flavored: true)

    obj["file_name"] = file_name
    obj["title"] = title
    obj["creation_datetime"] = created_datetime
    obj["updated_datetime"] = post.updated
    obj["content"] = content
    obj["categories"] = "[" + post.categories.join(", ").delete('\:') + "]"
    posts.push(obj)
  end
  return posts
end

def write_posts(posts, verbose)
  Dir.mkdir("_posts") unless File.directory?("_posts")

  total = posts.length, i = 1
  posts.each do |post|
    file_name = "_posts/".concat(post["file_name"])
    header = %{---
title: #{post["title"]}
layout: post
date: #{post["creation_datetime"]}
updated: #{post["updated_datetime"]}
comments: false
categories: #{post["categories"]}
---
}
    File.open(file_name, "w+") {|f|
      f.write(header)
      f.write(post["content"])
      f.close
    }

    if verbose
      puts "  [#{i}/#{total[0]}] Written post #{file_name}"
      i += 1
    end
  end
end

def main
  options = {}
  opt_parser = OptionParser.new do |opt|
    opt.banner = "Usage: ./blogspot_to_jekyll.rb FEED_URL [OPTIONS]"
    opt.separator ""
    opt.separator "Options"

    opt.on("-v", "--verbose", "Print out all.") do
      options[:verbose] = true
    end
  end

  opt_parser.parse!

  if ARGV[0]
    feed_url = ARGV.first
  else
    puts opt_parser
    exit()
  end

  puts "Fetching feed #{feed_url}..."
  feed = Feedjira::Feed.fetch_and_parse(feed_url)

  puts "Parsing feed..."
  posts = parse_post_entries(feed, options[:verbose])

  puts "Writing posts to _posts/..."
  write_posts(posts, options[:verbose])

  puts "Done!"
end

main()

run.sh

#!/bin/bash
#
# This is an example of how to run it, you need to put the size otherwise it will load just 25 posts and 
# have a max of 500 posts every time 
# http://too-clever-by-half.blogspot.kr/2011/12/blog-feed-500-post-limit-for-more-than.html
#
ruby blogspot_to_jekyll.rb http://<blog>.com/feeds/posts/default?max-results=200