Ruby script to convert google docs into markdown

doc-to-md.rb

# Usage ruby doc-to-md.rb input.html > output.md
# Make sure to install reverse_markdown: gem install reverse_markdown

require 'reverse_markdown'
require 'cgi'

input = ARGF.read

BOLD_CLASSES = input.scan(/\.(\w+)\s*\{[^}]*font-weight:\s*700[^}]*\}/).flatten

module ReverseMarkdown
  module Converters
    class Span < Base
      def convert(node, state = {})
        content = treat_children(node, state.merge(already_strong: true))
        classes = node.attributes['class']&.value.to_s.split(' ')

        if content.strip.empty? || state[:already_strong]
          ""
        elsif (classes & BOLD_CLASSES).any?
          "#{content[/^\s*/]}**#{content.strip}**#{content[/\s*$/]}"
        else
          content
        end
      end

      def disabled?
        !enabled?
      end
    end

    class PreventEmptyLinks < A
      def convert(node, state = {})
        content = treat_children(node, state)
        node['href'] = CGI.unescape(node['href'].gsub!(/https:\/\/www\.google\.com\/url\?q=([^&]+)[^\)]+/, '\1').to_s)
        content == "&nbsp;" ? '  ' : " #{super}"
      end
    end

    register :span, Span.new
    register :a, PreventEmptyLinks.new
  end
end

puts ReverseMarkdown.convert input