爬虫 💋

spider.rb

require 'open-uri'
require 'nokogiri'
require 'capybara/poltergeist'

url = "http://mp.weixin.qq.com/s?__biz=MzA3NDMyOTcxMQ==&mid=209031164&idx=1&sn=606c865bfbc982029b2b71731304d7bb&3rd=MzA3MDU4NTYzMw==&scene=6#rd"

Capybara.default_driver = :poltergeist
Capybara.run_server = false

doc = Nokogiri::HTML(open(url))

puts text = doc.css("#media").at('script').text

regexp = /cover = "(.*?)"/ =~ text

puts "*" *100

puts $1

# Configure Poltergeist to not blow up on websites with js errors aka every website with js
# See more options at https://github.com/teampoltergeist/poltergeist#customization
Capybara.register_driver :poltergeist do |app|
  Capybara::Poltergeist::Driver.new(app, js_errors: false)
end

# Configure Capybara to use Poltergeist as the driver
Capybara.default_driver = :poltergeist

browser = Capybara.current_session

browser.visit url

# browser.save_and_open_page #保存网页

# media = browser.all('#media')


img = browser.find("#media").find('img')['src']

puts img