siers/update.rb

## update.rb
#!/usr/bin/env ruby

require 'rubygems'
require 'pry'
require 'nokogiri'
require 'mini_magick'
require 'digest'

W = 412
H = 600

def template(text)
  <<-HTML
  <!DOCTYPE html>
  <html>
  <head>
    <title>saraksts</title>
    <meta charset="utf-8" />
  </head>
  <style>
  img {
    float: left;
    padding: 10px;
    display: block;
    border-left: 2px dotted #666;
    width: #{ W / 3 }px;
    height: #{ H / 3 }px;
  }
  img:first-child { border: none; }
  img:hover { position: relative; width: #{ W + 20 }px; height: #{ H + 20 }px; }
  </style>
  <body>
  #{ text }
  </body>
  </html>
  HTML
end

def inform(text)
  yield
ensure
  puts text
end

def curl(link)
  inform "<-- Downloading #{ link }" do
    %x{curl 2>/dev/null #{ link }}
  end
end

def parse(link)
  inform "--> Parsing #{ link }" do
    Nokogiri::HTML::Document.parse(curl link)
  end
end

def get_table(opts)
  filename, link = opts.first
  image = nil
  inform "::  Searching for table's image" do
    image = parse(link).css('p img').first.attributes["src"].value
  end

  crop(image, filename)
  filename
end

def crop(from, to)
  inform "::  Cropping" do
    image = MiniMagick::Image.open(from)
    image.crop("#{ W }x#{ H }+0+425")
    inform "--> Saving to #{ to }" do
      image.write(to)
    end
  end
end

doc   = parse('http://www.rskola.lv')
days  = doc.search("[text()*='diena']").select { |n| n.children.first.text =~ /[A-Z][a-z]+diena/ }
pages = days.map { |n| [n.children.first.text, n.attributes["href"].value] }

html = ''
pages.each do |day, page|
  filename = "#{ day }-#{ Digest::MD5.hexdigest(page) }.jpg"
  get_table(filename => page) unless File.exists?(filename)

  html << "<img src='#{ filename }' alt='#{ day }' />"
end

inform "--> Generating HTML" do
  File::open('index.html', 'w').write(template(html))
end
	#!/usr/bin/env ruby

	require 'rubygems'
	require 'pry'
	require 'nokogiri'
	require 'mini_magick'
	require 'digest'

	W = 412
	H = 600

	def template(text)
	<<-HTML
	<!DOCTYPE html>
	<html>
	<head>
	<title>saraksts</title>
	<meta charset="utf-8" />
	</head>
	<style>
	img {
	float: left;
	padding: 10px;
	display: block;
	border-left: 2px dotted #666;
	width: #{ W / 3 }px;
	height: #{ H / 3 }px;
	}
	img:first-child { border: none; }
	img:hover { position: relative; width: #{ W + 20 }px; height: #{ H + 20 }px; }
	</style>
	<body>
	#{ text }
	</body>
	</html>
	HTML
	end

	def inform(text)
	yield
	ensure
	puts text
	end

	def curl(link)
	inform "<-- Downloading #{ link }" do
	%x{curl 2>/dev/null #{ link }}
	end
	end

	def parse(link)
	inform "--> Parsing #{ link }" do
	Nokogiri::HTML::Document.parse(curl link)
	end
	end

	def get_table(opts)
	filename, link = opts.first
	image = nil
	inform ":: Searching for table's image" do
	image = parse(link).css('p img').first.attributes["src"].value
	end

	crop(image, filename)
	filename
	end

	def crop(from, to)
	inform ":: Cropping" do
	image = MiniMagick::Image.open(from)
	image.crop("#{ W }x#{ H }+0+425")
	inform "--> Saving to #{ to }" do
	image.write(to)
	end
	end
	end

	doc = parse('http://www.rskola.lv')
	days = doc.search("[text()*='diena']").select { \|n\| n.children.first.text =~ /[A-Z][a-z]+diena/ }
	pages = days.map { \|n\| [n.children.first.text, n.attributes["href"].value] }

	html = ''
	pages.each do \|day, page\|
	filename = "#{ day }-#{ Digest::MD5.hexdigest(page) }.jpg"
	get_table(filename => page) unless File.exists?(filename)

	html << "<img src='#{ filename }' alt='#{ day }' />"
	end

	inform "--> Generating HTML" do
	File::open('index.html', 'w').write(template(html))
	end