Open source books @ Kindle

I finally bought my Kindle Paperwhite (which is jaibroken btw). The cool thing about Kindle that it has community that provides a lot of good stuff:

I used last one, kindlefodder, to write recipes for books The Little Book on CoffeeScript and Smooth CoffeeScript

Kindlefodder takes a lot of routine work to itself.

First, I specified document’s info, including cover:

def document 
  # download cover image
  if !File.size?("cover.gif")
    `curl -s '' > cover.jpg`
    run_shell_command "convert cover.jpg -type Grayscale -resize '400x300>' cover.gif"
  # book's info
    'title' => 'The Little Book on CoffeeScript',
    'author' => 'Alex MacCaw',
    'cover' => 'cover.gif',
    'masthead' => nil

Then, we are fetching first page (with TOC) and formatting YAML with articles list:

def get_source_files
  # fetch first page (with TOC)
  @start_url = ""
  @start_doc = Nokogiri::HTML run_shell_command("curl -s #{@start_url}")

  # create sections.yml
  sections = [{
    }]"#{output_dir}/sections.yml", 'w') {|f|
    f.puts sections.to_yaml

Resulted sections.yml will look like:

- :title: Main
  - :title: Introduction
    :path: articles/01_introduction.html
  - :title: Syntax
    :path: articles/02_syntax.html
  - :title: Classes
    :path: articles/03_classes.html
  - :title: Idioms
    :path: articles/04_idioms.html
  - :title: Compiling
    :path: articles/05_compiling.html
  - :title: Applications
    :path: articles/06_applications.html
  - :title: The Bad Parts
    :path: articles/07_the_bad_parts.html

So, it has one section (Main) and a lot of articles. More complex books could have multiple sections, of course.

Let’s take a look on extract_articles method:

def extract_articles
  # iterating over Table of Contents and extracting articles'ol.pages li a').map do |o|
    title = o.inner_text

    FileUtils::mkdir_p "#{output_dir}/articles"

      title: title,
      path: save_article_and_return_path(o[:href])

And save_article_and_return_path method, which fetches actual article, cleans it, saves it and returns saved article path:

def save_article_and_return_path href, filename=nil
  path = filename || "articles/" + href.sub(/^\//, '').sub(/\/$/, '').gsub('/', '.')
  # fetching article
  full_url = @start_url + href.sub(/^\//, '')
  html = run_shell_command "curl -s #{full_url}"
  # cleaning article
  article_doc = Nokogiri::HTML html
  b =".back")
  b.remove if b
  # saving article
  res ='#content').inner_html"#{output_dir}/#{path}", 'w') {|f| f.puts res}

I’m using beautiful Nokogiri ruby library for chopping of HTML here.

You don’t have to care about images, .mobi format and stuff like that, Kindlefodder does it for you.

So, resulted recipes:

blog comments powered by Disqus