manually-migrate.rb

require 'open-uri'
require 'uri'
require 'fileutils'
require 'nokogiri'
require 'optparse'
require 'base64'
require 'json'


# Adds support for optional flags to the script
@options = {
  :content_html_script => false,
  :copy_to_clipboard => true,
  :original_url => nil,
  :new_url => nil,
  :base64_img => false,
  :update => false,
  :parse_local => false,
  :json_encode => false
}

OptionParser.new do |opts|
  opts.banner = "Usage: manually-migrate.rb [options]"

  opts.on("-o", "--content-html-script [true]", TrueClass, "Adds script to paste HTML inside content editor (default: false)") do |v|
    @options[:content_html_script] = v.to_s.downcase == 'true' || v.to_s.downcase == ''
  end

  opts.on("-c", "--copy-to-clipboard [true]", TrueClass, "Copies console script to clipboard (default: true)") do |v|
    @options[:copy_to_clipboard] = v.to_s.downcase == 'true' || v.to_s.downcase == ''
  end

  opts.on("-u", "--original-url [url]", "Original url to migrate without user input (default: user input)") do |v|
    @options[:original_url] = v
  end

  opts.on("-n", "--new-url [url]", "New url to open editing admin page (default: nothing)") do |v|
    @options[:new_url] = v.to_s.downcase == 'true' || v.to_s.downcase == ''
    if @options[:new_url] != true
      @options[:new_url] = v
    end
  end

  opts.on("-b", "--base64-img [true]", TrueClass, "Encode <img> tags as base64 data (default: false)") do |v|
    @options[:base64_img] = v.to_s.downcase == 'true' || v.to_s.downcase == ''
  end

  opts.on("-e", "--update [true]", TrueClass, "Updates migration script (default: false)") do |v|
    @options[:update] = v.to_s.downcase == 'true' || v.to_s.downcase == ''
  end

  opts.on("-p", "--parse-local [true]", TrueClass, "Parses pasted content (default: false)") do |v|
    @options[:parse_local] = v.to_s.downcase == 'true' || v.to_s.downcase == ''
  end

  opts.on("-j", "--json-encode [true]", TrueClass, "Generate notes in json format (default: false)") do |v|
    @options[:json_encode] = v.to_s.downcase == 'true' || v.to_s.downcase == ''
  end
end.parse!

if @options[:update] == true
  puts 'Updating migration script'
  open('https://raw.githubusercontent.com/dave11mj/dh-migration/master/manually-migrate.rb') do |f|
    File.open('manually-migrate.rb','w') do |file|
      file.puts f.read
    end
  end
  exit
end

if @options[:parse_local] == true
    print "html to parse     "
    original_html = File.open("parse-option.html")
    page = Nokogiri::HTML(original_html.read)
    FileUtils::mkdir_p "./downloads/parse_local"
    Dir.chdir "./downloads/parse_local"
end

unless @options[:parse_local] == true

  if @options[:original_url] == nil
    print "Original url: "
    original_url = STDIN.gets.strip
  else
    original_url = @options[:original_url]
  end

  # Asks for New URL if new url option is true
  if @options[:new_url] == true
    print "New url: "
    @options[:new_url] = STDIN.gets.strip
  end

  uri = URI::parse(original_url)

  FileUtils::mkdir_p "./downloads#{uri.path}"

  Dir.chdir "./downloads#{uri.path}"

  current_folder = uri.path.split('/').last

  html = open(original_url)
  page = Nokogiri::HTML(html.read)
  page.encoding = 'utf-8'

end


unless page.css("title").to_s == ''
  @page_head_title = page.css("title")[0].text
end

@page_head_title = ''
unless page.css("h1").to_s == ''
  @page_head_title = page.css("h1")[0].text
end

@page_head_description = page.at("meta[name=description]")

if @page_head_description != nil
  @page_head_description = @page_head_description['content']
elsif page.css("h2").to_s != ''
  @page_head_description = page.css("h2")[0].text
else
  @page_head_description = ''
end

# Removes inline styles
page.xpath('@style|.//@style').remove

# Removes align attributes
page.xpath('@align|.//@align').remove

@page_body_title = ''
unless page.css("h1").to_s == ''
  @page_body_title = page.css("h1")[0].text
end

@page_body_description = ''
unless page.css("h2").to_s == ''
  page_body_description = page.css("h2")[0].text
end

if @options[:parse_local]
  page_body_html = page.to_s.gsub!(/[\n\t\r]+/, '')
else
  page_body_html = page.css('span#HTML_CONTENT')[0].to_s.gsub!(/[\n\r]+/, '')
end

# Removes <script> tags
page.css('script').remove

downloaded_images = []

unless page_body_description == nil
  # Format Phone Numbers on Description
  page_body_description.gsub!(/(?:\d{1}[[:space:]])?\(?(\d{3})\)?-?[[:space:]]?(\d{3})-?[[:space:]]?(\d{4})/, '<strong><a href="tel:+1-\1-\2-\3">\1.\2.\3</a></strong>')

  # Exception regex for phone numbers already with dot format
  page_body_description.gsub!(/(?:\d{1}\s)?\(?(\d{3})\)?\.(\d{3})\.(\d{4})/, '<strong><a href="tel:+1-\1-\2-\3">\1.\2.\3</a></strong>')
end

unless page_body_html == nil

  # Remove IDs
  # page_body_html.gsub!(/id="?[^"\s]*"?/, '')

  if @options[:parse_local] != true
    # Remove <span> and <div> tags
    page_body_html.gsub!(/(<\/?(span|div)[^>]*>)/, '')
  end

  # Validates HTML for unclosed tags
  page_body_html = Nokogiri::HTML::DocumentFragment.parse(page_body_html).to_s

  # Remove line breaks
  page_body_html.gsub!(/[\n\r]+/, '')

  # Replace h1 and h2 with h3
  page_body_html.gsub!(/<(\/?)h[12][^>]*>/, '<\1h3>')

  # Remove Tel Links
  page_body_html.gsub!(/<a href="(tel:|\d)[^>]+>(.+?)<\/a>/, '\2')

  # Normalizes phone numbers to use dot format
  page_body_html.gsub!(/(?:\d{1}[[:space:]])?\(?(\d{3})\)?-?[[:space:]]?(\d{3})-?[[:space:]]?(\d{4})/, '\1.\2.\3')

  # Wraps phone numbers on tel links and strong tags
  page_body_html.gsub!(/(?:\d{1}\s)?\(?(\d{3})\)?\.(\d{3})\.(\d{4})/, '<strong><a href="tel:+1-\1-\2-\3">\1.\2.\3</a></strong>')

  # Remove Updated or Rev number
  page_body_html.gsub!(/\((Updated|Rev)\.? \d+\.?\)/, '')

  # Replaces any Find a doctor link with the correct url
  page_body_html.gsub!(/<a[^>]*href="[^"]*(?:\b[Ff]ind\b|\bourdoctors\b)[^"]*"[^>]*>([^<]*[Dd]octor[^<]*<\/a>)/, '<a href="~/link.aspx?_id=33D04E30953B4568809EDDDB52367C62&amp;_z=z" target="_blank">\1')

  # Find and save images
  images = page_body_html.scan(/<img.*?>/)
  unless images == nil or @options[:parse_local] == true
    images.each.with_index(1) do |image, index|
      src = image.match(/(?<=src=")[^"]+/).to_s

      if src.match(/https?:\/\//)
        image_url = src
      else
        image_url = "#{uri.scheme}://#{uri.host}#{src}"
      end

      open(image_url) do |f|
        if @options[:base64_img]
          page_body_html.sub!(/src=["']?(?!data)[^"'\s>]*["']?(\s|>)/, "src='data:image/jpeg;base64, #{Base64.encode64(f.read).gsub(/\n/, '')}'\\1")
        else
          File.open("inline-#{current_folder}-photo-#{index}.jpg","wb") do |file|
            file.puts f.read
          end

          downloaded_images << "file: inline-#{current_folder}-photo-#{index}.jpg — alt: #{(image.match(/(?<=alt=")[^"]+/) || 'none')}\n"
        end
      end
    end
  end

  # Remove Image tags
  unless @options[:base64_img]
    page_body_html.gsub!(/<img.*?>/, '')
  end

  # Removes Duplicate strong tags
  page_body_html.gsub!(/<(strong)[^>]*>\s*?<\1[^>]*>(.*?)<\/\1>\s*?<\/\1>/, '<\1>\2</\1>')

  if @options[:parse_local] != true
    # Remove Empty tags that are not iframe, or wrapping an image
    page_body_html.gsub!(/<(?!iframe)([^>\s]+)([^>]+)?>([[:space:]]|&nbsp;|<(?!img)[^>]+>)*<\/\1>/, '')
  end

  # Wraps any block level text without tags on <p>
  page_body_html.gsub!(/(<\/(p|ul|h3)>)([^<]+?)(<(p|ul|h3)[^>]*>)/, '\1<p>\3</p>\4')

end

styles = "<style>"\
          ".manually-migrated ul, .manually-migrated ol { padding-left: 40px; } "\
          ".manually-migrated img { display:block; float: right; clear: right; margin: 0 0 20px 20px; } "\
          "@media (max-width: 750px) { .manually-migrated img { display:block; float: none; clear: both; margin: 20px auto; max-width: 100%; } } "\
          "</style>"


page_body_description_html = ""
unless page_body_description == ''
  page_body_description_html = "<p>#{page_body_description}</p>"
end

@main_content_html = "#{styles} <div class='manually-migrated'>#{page_body_description_html}#{page_body_html}</div>"

def console_script_generator(new_url = false)

  jQueryFind = (new_url) ? "jQuery('#scPageExtendersForm iframe').contents().find" : "jQuery"

  tmp_script = ""
  br = (@options[:json_encode] == true) ? '' : "\n"
  # Script used to open 'edit html' editor and paste content html inside of it
  if @options[:content_html_script]
    console_content_html_script = "#{jQueryFind}(\"#Section_Content\").next().find(\".scContentButton:contains('Edit HTML')\").eq(0).trigger('click'); "\
    "(function updateIframe() { "\
        "var $contentHtml = jQuery('#jqueryModalDialogsFrame').contents().find('#scContentIframeId0').contents().find('textarea'); "\
        "var attempts = 1; "\
        "if($contentHtml.length < 1 && attempts <= 60) { "\
            "setTimeout(updateIframe, 500); "\
            "console.log('attempts to fill html', attempts); "\
            "attempts++; "\
        "} else if (attempts <= 60) { "\
          "console.log('Success ! Pasting HTML'); "\
          "$contentHtml.val('#{@main_content_html.gsub(/'/, "\\\\'")}');"\
        "} else { "\
          "console.log('Waited for too long.. Stopping attempts'); "\
        "} "\
    "})();\n"
  else
    console_content_html_script = ""
  end

  tmp_script = "#{jQueryFind}(\".scEditorFieldLabel:contains('Title'), .scEditorFieldLabel:contains('Header')\").next().children('input').val('#{@page_body_title.gsub(/'/, "\\\\'")}');\n"\
                  "#{jQueryFind}(\".scEditorFieldLabel:contains('PageHeadTitle:')\").next().children('input').val('#{@page_head_title.gsub(/'/, "\\\\'")}');#{br}"\
                  "#{jQueryFind}(\".scEditorFieldLabel:contains('PageHeadDescription')\").next().children('input').val('#{@page_head_description.gsub(/'/, "\\\\'")}');#{br}"\
                  "#{jQueryFind}(\".scEditorFieldLabel:contains('Display name')\").next().children('input').val('#{@page_body_title.gsub(/'/, "\\\\'")}');#{br}"\
                  "#{jQueryFind}(\".scEditorFieldLabel:contains('PageAddToSitemap')\").prev().children('input').prop('checked', true);#{br}"\
                  "#{jQueryFind}(\".scEditorFieldLabel:contains('PageShowInSearch')\").prev().children('input').prop('checked', true);#{br}"\
                  "#{console_content_html_script}"

  if new_url
    tmp_script << "jQuery('#main-nav, #mobile-nav').css('visibility', 'hidden');"
  end

  return tmp_script
end

console_script = ""

if @options[:parse_local] == false
  console_script = console_script_generator
end

new_url_console_script = ""
new_url_notes = ""

edit_page_url = ""
edit_page_notes = ""

if @options[:new_url] != nil
  new_url_console_script = console_script_generator(true)
  new_url_notes = "\n\n###New URL Console Script\n#{new_url_console_script}\n"

  new_url_uri = URI::parse(@options[:new_url])
  edit_page_url = "https://slot2.dev.dignityhealth.org/sitecore/content/Service%20Areas#{new_url_uri.path}?sc_mode=edit&sc_ce=1"
  edit_page_notes = "\n\n###Edit Page URL\n#{edit_page_url}\n"
end

if @options[:json_encode] == true
  json_notes = {
    :original_url => original_url,
    :edit_page_url => edit_page_url,
    :page_head_title => @page_head_title,
    :page_head_description => @page_head_description,
    :page_body_title => @page_body_title,
    :main_content_html => @main_content_html,
    :downloaded_images => downloaded_images.join(''),
    :console_script => "#{console_script}#{new_url_console_script}"
  }

  puts JSON.generate(json_notes)

  File.open("notes.json", 'w') { |file| file.write(JSON.pretty_generate(json_notes)) }
else
  notes = "###Original URL\n"\
          "#{original_url}#{edit_page_notes}\n\n"\
          "###PageHeadTitle\n"\
          "#{@page_head_title}\n\n"\
          "###PageHeadDescription\n"\
          "#{@page_head_description}\n\n"\
          "###Content Title\n"\
          "#{@page_body_title}\n\n"\
          "###Content HTML\n"\
          "#{@main_content_html}\n\n"\
          "###Downloaded Images\n"\
          "#{downloaded_images.join('')}\n"\
          "###Console Script\n"\
          "#{console_script}#{new_url_notes}\n"

  puts "\n#{notes}\n"

  File.open("notes.md", 'w') { |file| file.write(notes) }
end

# Copies to clipboard on Macs
def pbcopy(input)
  str = input.to_s
  IO.popen('pbcopy', 'w') { |f| f << str }
  str
end

if @options[:copy_to_clipboard] && @options[:json_encode] == false
  begin
    if @options[:new_url] != nil
      pbcopy(new_url_console_script)
      puts "Copied New URL Console Script to clipboard\n\n"
    else
      pbcopy(console_script)
      puts "Copied Console Script to clipboard\n\n"
    end
  rescue
    puts "Done\n\n"
  end
elsif @options[:json_encode] == false
  puts "Done\n\n"
end

# Open Admin Edit page if new url is provided
if @options[:new_url] != nil && @options[:parse_local] == false && @options[:json_encode] == false
  begin
    require 'launchy'
    Launchy.open(edit_page_url)
  rescue
    puts 'It seems you are missing a dependency. To use the --new-url flag'
    puts 'Please run `sudo gem install launchy`'
  end
end