我试图开发一个简单的网页刮板。我想提取没有HTML代码的文本。它适用于普通HTML,但不适用于JavaScript代码添加文本的某些页面。

例如,如果一些JavaScript代码添加了一些文本,我不能看到它,因为当我调用:

response = urllib2.urlopen(request)

我得到了原始文本而没有添加的文本(因为JavaScript是在客户端执行的)。

所以,我正在寻找一些解决这个问题的想法。


当前回答

EDIT 2021年9月:phantomjs也不再维护

EDIT 30/Dec/2017:这个答案出现在谷歌搜索的顶部结果中,所以我决定更新它。老答案仍然在最后。

dryscape不再维护,开发人员推荐的库dryscape仅适用于Python 2。我发现使用Selenium的python库和Phantom JS作为web驱动程序足够快,也很容易完成工作。

一旦你安装了Phantom JS,确保phantomjs二进制文件在当前路径下可用:

phantomjs --version
# result:
2.1.1

#例子 为了给出一个例子,我用下面的HTML代码创建了一个示例页面。(链接):

<!DOCTYPE html>
<html>
<head>
  <meta charset="utf-8">
  <title>Javascript scraping test</title>
</head>
<body>
  <p id='intro-text'>No javascript support</p>
  <script>
     document.getElementById('intro-text').innerHTML = 'Yay! Supports javascript';
  </script> 
</body>
</html>

没有javascript,它说:不支持javascript和javascript:耶!支持javascript

#抓取没有JS支持:

import requests
from bs4 import BeautifulSoup
response = requests.get(my_url)
soup = BeautifulSoup(response.text)
soup.find(id="intro-text")
# Result:
<p id="intro-text">No javascript support</p>

#抓取与JS支持:

from selenium import webdriver
driver = webdriver.PhantomJS()
driver.get(my_url)
p_element = driver.find_element_by_id(id_='intro-text')
print(p_element.text)
# result:
'Yay! Supports javascript'

你也可以使用Python库dryscraping来抓取javascript驱动的网站。

#抓取与JS支持:

import dryscrape
from bs4 import BeautifulSoup
session = dryscrape.Session()
session.visit(my_url)
response = session.body()
soup = BeautifulSoup(response)
soup.find(id="intro-text")
# Result:
<p id="intro-text">Yay! Supports javascript</p>

其他回答

Pyppeteer

你可以考虑Pyppeteer,它是Chrome/Chromium驱动程序前端的Python移植版本。

下面是一个简单的例子,展示了如何使用pyppeterer动态地访问被注入到页面中的数据:

import asyncio
from pyppeteer import launch

async def main():
    browser = await launch({"headless": True})
    [page] = await browser.pages()

    # normally, you go to a live site...
    #await page.goto("http://www.example.com")
    # but for this example, just set the HTML directly:
    await page.setContent("""
    <body>
    <script>
    // inject content dynamically with JS, not part of the static HTML!
    document.body.innerHTML = `<p>hello world</p>`; 
    </script>
    </body>
    """)
    print(await page.content()) # shows that the `<p>` was inserted

    # evaluate a JS expression in browser context and scrape the data
    expr = "document.querySelector('p').textContent"
    print(await page.evaluate(expr, force_expr=True)) # => hello world

    await browser.close()

asyncio.get_event_loop().run_until_complete(main())

请参阅pyppeterer的参考文档。

Selenium是抓取JS和Ajax内容的最佳工具。

查看这篇文章,了解如何使用Python从web中提取数据

$ pip install selenium

然后下载Chrome webdriver。

from selenium import webdriver

browser = webdriver.Chrome()

browser.get("https://www.python.org/")

nav = browser.find_element_by_id("mainnav")

print(nav.text)

容易,对吧?

Playwright-Python

还有一种选择是剧作家- Python,它是微软剧作家(本身是受木偶大师影响的浏览器自动化库)到Python的移植。

下面是选择一个元素并抓取它的文本的最小示例:

from playwright.sync_api import sync_playwright

with sync_playwright() as p:
    browser = p.chromium.launch()
    page = browser.new_page()
    page.goto("http://whatsmyuseragent.org/")
    ua = page.query_selector(".user-agent");
    print(ua.text_content())
    browser.close()

You'll want to use urllib, requests, beautifulSoup and selenium web driver in your script for different parts of the page, (to name a few). Sometimes you'll get what you need with just one of these modules. Sometimes you'll need two, three, or all of these modules. Sometimes you'll need to switch off the js on your browser. Sometimes you'll need header info in your script. No websites can be scraped the same way and no website can be scraped in the same way forever without having to modify your crawler, usually after a few months. But they can all be scraped! Where there's a will there's a way for sure. If you need scraped data continuously into the future just scrape everything you need and store it in .dat files with pickle. Just keep searching how to try what with these modules and copying and pasting your errors into the Google.

如果你以前曾经使用过python的Requests模块,我最近发现开发人员创建了一个名为Requests- html的新模块,现在它也有呈现JavaScript的能力。

你也可以访问https://html.python-requests.org/来了解更多关于这个模块的信息,或者如果你只对呈现JavaScript感兴趣,那么你可以访问https://html.python-requests.org/?#javascript-support来直接学习如何使用该模块使用Python来呈现JavaScript。

从本质上讲,一旦你正确安装了Requests-HTML模块,下面的例子,在上面的链接中显示,展示了你如何使用这个模块来抓取一个网站,并呈现网站中包含的JavaScript:

from requests_html import HTMLSession
session = HTMLSession()

r = session.get('http://python-requests.org/')

r.html.render()

r.html.search('Python 2 will retire in only {months} months!')['months']

'<time>25</time>' #This is the result.

我最近从YouTube上的一个视频中了解到这一点。点击这里!观看YouTube上演示该模块如何工作的视频。