Skip to content

Instantly share code, notes, and snippets.

@kennym
Created July 30, 2011 18:14
Show Gist options
  • Save kennym/1115810 to your computer and use it in GitHub Desktop.
Save kennym/1115810 to your computer and use it in GitHub Desktop.
Migrate your blogger blog posts to jekyll.
#!/usr/bin/env ruby
#
# Convert blogger (blogspot) posts to jekyll posts
#
# Basic Usage
# -----------
#
# ./blogger_to_jekyll.rb feed_url
#
# where `feed_url` can have the following format:
#
# http://{your_blog_name}.blogspot.com/feeds/posts/default
#
# Requirements
# ------------
#
# * feedzirra: https://github.com/pauldix/feedzirra
#
# Notes
# -----
#
# * Make sure Blogger shows full output of article in feeds.
# * Commenting on migrated articles will be set to false by default.
include Config
require 'rubygems' if CONFIG['host_os'].start_with? "darwin"
require 'feedzirra'
require 'date'
require 'optparse'
def parse_post_entries(feed, verbose)
posts = []
feed.entries.each do |post|
obj = Hash.new
created_datetime = post.updated
creation_date = Date.parse(created_datetime.to_s)
title = post.title
file_name = creation_date.to_s + "-" + title.split(/ */).join("-").delete('\/') + ".html"
content = post.content
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(" ")
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 = %{---
layout: post
title: #{post["title"]}
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: ./blogger_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 = Feedzirra::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()
@kennym
Copy link
Author

kennym commented Apr 23, 2012

I just merged your changes. Sorry, that I couldn't figure out a way to keep the credits, but your work is most appreciated! Thank you, sir

@bebraw
Copy link

bebraw commented Apr 23, 2012

No probs. Thanks. :)

@bebraw
Copy link

bebraw commented Apr 23, 2012

Hi, one more thing... I noticed it is better to use "created_datetime = post.published" instead of "created_datetime = post.updated". Might want to change that in your gist. You might want to store the updated time in YAML front matter btw. Probably doesn't hurt.

@kennym
Copy link
Author

kennym commented Apr 23, 2012

You might want to store the updated time in YAML front matter btw. Probably doesn't hurt.

You lost me. YAML?

@bebraw
Copy link

bebraw commented Apr 24, 2012

See this. It's just that little snippet in front of a page.

I updated my Gist to contain the change mentioned above. Feel free to merge.

@kennym
Copy link
Author

kennym commented Apr 24, 2012

Merged. :)

@basav
Copy link

basav commented Aug 6, 2012

Thanks for this gist. Works

@jorgeramirez
Copy link

After changing Config to RbConfig it worked. Thanks!

@coolaj86
Copy link

FYI, if you'd also like to migrate the comments: http://blog.coolaj86.com/articles/migrate-from-blogger-to-ruhoh-with-proper-redirects.html

This requires changing your template (explained in the walkthrough) and exporting a backup of your blog (also explained in the walkthrough)

@mpenkov
Copy link

mpenkov commented Nov 1, 2013

Thanks for this script! Everything worked well, except the handling of colon characters in the title. They make Jekyll fall over and die, for some reason. Relevant: jekyll/jekyll#549

@nikhil91
Copy link

thanks:)

@shiv4289
Copy link

Hi Kennym,

Worked for me. Works like a charm. But I had some trouble because the feedzirra module is now renamed to feedjira.
You need to update the script to show that. I did the same and I could do the import then.

I have one question though: I lost the comments I had on blogger in the process. How do I migrate the comments?

@raphink
Copy link

raphink commented Aug 13, 2014

+1 for feedjira

@DrizzlingCattus
Copy link

thanks for script!
when I rename feedzirra to feedjira, it work.
but feeds/posts/default option parse only some part of all my posts.
so, I change feeds/posts/default to feeds/posts/default?max-results=100 and it parse all my post.

I link about parsing all post.
http://too-clever-by-half.blogspot.kr/2011/12/blog-feed-500-post-limit-for-more-than.html

@danielgomezrico
Copy link

I´m getting blogspot_to_jekyll.rb:25:in

': Use RbConfig instead of obsolete and deprecated Config.
blogspot_to_jekyll.rb:27:in <main>': uninitialized constant CONFIG (NameError) after rename it to feedjira, why?

@lolobosse
Copy link

@danielgomezrico I got the same problem and solved it on my fork @kennym if you want, you can update yours from my code 😄 👍

@mwschwehr
Copy link

Hi kennym,
Great code. Worked for me on Yosemite with some minor changes.

I removed the deprecated CONFIG call. I think rubygems is now required for El Capitan anyway.

Feedzirra is now called feedjira, so I made the appropriate changes in the code.

After these two minor changes, the code worked perfectly 10.11.3

Feel free to do a pull and merge. In my commit message, i inadvertently stated I was updated for Yosemite. This is my first fork, edit, and push of code on Git, and my first time working in Ruby.

@savita20
Copy link

savita20 commented May 3, 2018

Even I need help to migrate my blog 'https://shindesavita87.blogspot.co.uk' from blogspot to GitHub blog. Can you suggest me is it doable and if yes and how can we do that?

@yuceltoluyag
Copy link

main': undefined method fetch_and_parse' for Feedjira::Feed:Class (NoMethodError)

@kennym
Copy link
Author

kennym commented Jan 24, 2020

@yuceltoluyag - I'm pretty sure this script might need some updates after 9 years :-D

@yuceltoluyag
Copy link

@yuceltoluyag - I'm pretty sure this script might need some updates after 9 years :-D

https://stackoverflow.com/questions/37371947/importing-my-blogger-blog-into-jekyll solved my problem =) ty for answer ;)

@rupeshtiwari
Copy link

I am getting this error <internal:C:/Ruby30-x64/lib/ruby/3.0.0/rubygems/core_ext/kernel_require.rb>:85:in require': cannot load such file -- feedzirra`

@RobbiNespu
Copy link

If anyone looking new changes. This is my fork https://gist.github.com/RobbiNespu/372571e4d271122ece3ee3a1830b4d26

@prabathbr
Copy link

If anyone looking new changes. This is my fork https://gist.github.com/RobbiNespu/372571e4d271122ece3ee3a1830b4d26

Thanks for the update. I have updated the code to fix few more errors which I have encountered.

My Fork: https://gist.github.com/prabathbr/0bb416b2dee7ed18d2a6fd3d8dd4b021

Updates:

  • added <require 'httparty'>
  • created "setup.sh" which will make a ruby environment to run the script in Ubuntu 20.04 LTS
  • fixed script error[Invalid argument @ rb_sysopen --- post name ---- (Errno::EINVAL)] when running on posts with invalid post names with ":" & "*"

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment