如何使用水豚獲取動態頁面內容

Question

嘗試從元素中獲取文本：

<div class="points-text" data-reactid=".2765swfgy68.1.2.2.1:$sony-xperia-z1-compact.2.0.4.0.0.1">29,367 points</div>

我猜網站使用Reactjs，即使使用Poltergeist驅動程序，水豚也無法獲得內容。 有什么解決方法嗎？

這是我的代碼：

require 'rubygems'
require 'capybara'
require 'capybara/poltergeist'


Capybara.default_driver = :poltergeist
Capybara.register_driver :poltergeist do |app|
  Capybara::Poltergeist::Driver.new(app, {js_errors: false})
end


class WebScraper
  include Capybara::DSL

  def get_page_data(url)
    visit(url)
    doc = Nokogiri::HTML(page.html)
    p doc.css('.points-text')
  end
end


scraper = WebScraper.new
puts scraper.get_page_data('http://versus.com/en/sony-xperia-z1-compact')

Answer 1

如果您已經使用Capybara進行了訪問，則無需使用Nokogiri解析html。

def get_page_data(url)
  visit(url)
  p find(:css, '.points-text').text      
end

將使用class points-text類在元素中顯示可見文本

Answer 2

此代碼適用於selenium和webkit，但不適用於poltergeist驅動程序。 從屏幕截圖中，我可以看到該網站上的javascript甚至沒有執行。

require 'capybara'
require 'capybara-webkit'
require 'capybara/poltergeist'

Capybara.default_driver = :poltergeist
# Capybara.default_driver = :selenium
# Capybara.default_driver = :webkit
Capybara.register_driver :poltergeist do |app|
  Capybara::Poltergeist::Driver.new(app, {js_errors: false})
end
Capybara::Webkit.configure do |config|
  config.allow_unknown_urls
end


class WebScraper
  include Capybara::DSL

  def get_page_data(url)
    visit(url)

    max_same_times = 3
    same_times = 0
    old_points = nil
    20.times do |i|
      doc = Nokogiri::HTML(page.html)
      # p doc.css('.points-text')
      points = doc.css('.points-text')
      # p [same_times, max_same_times]
      if same_times == max_same_times
        break
      end
      if points.length > 0
        points = points[0].text
        # p [old_points, points]
        if old_points == points
          same_times += 1
        else
          same_times = 0
        end
        old_points = points
      end
      page.save_screenshot("#{i}.png")
    end
    old_points
  end
end


scraper = WebScraper.new
puts scraper.get_page_data('http://versus.com/en/sony-xperia-z1-compact')