使用Selenium爬虫自动爬取广外表白墙

相信很多同学都会刷每期的表白墙，并期待从表白墙上找到自己的名字。那么作为信息学院的学生，有没有办法能自动获取到表白墙呢？

Selenium包可以用于自动控制网页，是一种很好的爬虫工具。

1. 首先我们布置环境

# coding: utf-8
import sys
from selenium import webdriver
from selenium.common.exceptions import NoSuchElementException
from selenium.webdriver.chrome.options import Options
import selenium
import json
import time
import requests

2. 我们使用chrome浏览器实现自动操作，首先需要下载针对chrome的webdriver驱动chromedriver，然后对浏览器进行设置。

chrome_options = webdriver.ChromeOptions()
prefs = {"profile.managed_default_content_settings.images":2} #设置不加载图片，可以加快网页加载速度
chrome_options.add_experimental_option("prefs",prefs)
driver = webdriver.Chrome("./chromedriver", chrome_options=chrome_options) #chromedriver的路径
# driver.set_page_load_timeout(10)
# driver.set_script_timeout(15)
driver.set_window_size(1366, 768) #设置浏览器打开的分辨率

3. 然后我们分析表白墙的网页

然后我们发现网页url为 “https://weixin.sogou.com/weixin?query=%E5%B9%BF%E5%A4%96%E8%A1%A8%E7%99%BD%E5%A2%99&type=2&page=2” ， 因此翻页比较简单，只需要更换page后面的数字即可（用click点击下方的翻页按钮也可以）。

那我们怎么点进去每一篇表白墙，获取里面的内容呢？思路就是：在外面的网页列表进行翻页，并获取每一遍表白墙的url，然后将url写入列表，传给下一个函数，然后下一个函数一个个打开这些网页，然后再获取里面的内容。

我们点击f12

我们使用xpath确定表白墙标题的url。xpath是一种基于html的定位方式，把html路径确定好就可以定位到具体的标签。然后将url写入列表。

def release(url, all_urls):
    driver.get(url)
    content_div = driver.find_elements_by_xpath(u"//ul[@class='news-list']//h3/a")
    hrefs = [e.get_attribute("href") for e in content_div]
    all_urls.extend(hrefs)
    time.sleep(3)

把获取到的url列表保存到json文件。

def save_state(all_urls, crawled_urls):
    with open("../data/state.json", "w+") as f:
        json.dump({
            "all": all_urls,
            "crawled": crawled_urls,
        },f)

读取url列表。

def load_state():
    with open("../data/state.json", "r") as f:
        d = json.load(f)
    all_urls = d["all"]
    crawled_urls = d["crawled"]
    return all_urls, crawled_urls

其中的意思是，将已爬取和未爬取的url分为两个列表，分别进行保存。

然后我们对每一篇表白墙的内容进行分析。

感觉表白墙的内容还是比较简单的，只需要获得section标签下的text内容即可。

def parse(url,all_urls):
    driver.get(url)
    data = driver.find_element_by_xpath(u"//div[@id='js_content']").text
    print(data)
    return data

我们将爬下来的表白墙保存。

def save2file(data, news_id):
    with open("../data/"+str(news_id)+".txt","w+") as f:
        f.write(data)
        f.close()

最后是主函数

def main():
    all_urls = []
    crawled_urls = []
    for i in range(10):
        url = "https://weixin.sogou.com/weixin?query=%E5%B9%BF%E5%A4%96%E8%A1%A8%E7%99%BD%E5%A2%99&type=2&page=" + str(i)
        release(url, all_urls)
        save_state(all_urls, crawled_urls)
        all_urls,crawled_urls = load_state()
    n = 1
    while len(all_urls) > 0:
        url = all_urls.pop(0)
        if url in crawled_urls:
            continue
        print("正在爬取：",url,)
        data = parse(url,all_urls)
        save2file(data, n)
        crawled_urls.append(url)
        save_state(all_urls, crawled_urls)
        print("已保存。")
        n += 1
        # except selenium.common.exceptions.NoSuchWindowException as ex2:
        #     print("爬虫终止")
        #     pass
        # except Exception as ex1:
        #     all_urls.append(url)
        #     save_state(all_urls, crawled_urls)
        #     print("出错啦！！！！！！",ex1)
        #     pass

main()

所有代码：

# coding: utf-8
import sys
from selenium import webdriver
from selenium.common.exceptions import NoSuchElementException
from selenium.webdriver.chrome.options import Options
import selenium
import json
import time
import requests



chrome_options = webdriver.ChromeOptions()
prefs = {"profile.managed_default_content_settings.images":2} #设置不加载图片，可以加快网页加载速度
chrome_options.add_experimental_option("prefs",prefs)
driver = webdriver.Chrome("./chromedriver", chrome_options=chrome_options) #chromedriver的路径
# driver.set_page_load_timeout(10)
# driver.set_script_timeout(15)
driver.set_window_size(1366, 768) #设置浏览器打开的分辨率

# global driver
# chrome_options = Options()
# chrome_options.add_argument('--headless')
# chrome_options.add_argument('--disable-gpu')
#
# driver = webdriver.Chrome(chrome_options=chrome_options)
#



def release(url, all_urls):
    driver.get(url)
    content_div = driver.find_elements_by_xpath(u"//ul[@class='news-list']//h3/a")
    hrefs = [e.get_attribute("href") for e in content_div]
    all_urls.extend(hrefs)
    time.sleep(3)




def parse(url,all_urls):
    driver.get(url)
    data = driver.find_element_by_xpath(u"//div[@id='js_content']").text
    print(data)
    return data

def save2file(data, news_id):
    with open("../data/"+str(news_id)+".txt","w+") as f:
        f.write(data)
        f.close()

def save_state(all_urls, crawled_urls):
    with open("../data/state.json", "w+") as f:
        json.dump({
            "all": all_urls,
            "crawled": crawled_urls,
        },f)

def load_state():
    with open("../data/state.json", "r") as f:
        d = json.load(f)
    all_urls = d["all"]
    crawled_urls = d["crawled"]
    return all_urls, crawled_urls

def main():
    all_urls = []
    crawled_urls = []
    for i in range(10):
        url = "https://weixin.sogou.com/weixin?query=%E5%B9%BF%E5%A4%96%E8%A1%A8%E7%99%BD%E5%A2%99&type=2&page=" + str(i)
        release(url, all_urls)
        save_state(all_urls, crawled_urls)
        all_urls,crawled_urls = load_state()
    n = 1
    while len(all_urls) > 0:
        url = all_urls.pop(0)
        if url in crawled_urls:
            continue
        print("正在爬取：",url,)
        data = parse(url,all_urls)
        save2file(data, n)
        crawled_urls.append(url)
        save_state(all_urls, crawled_urls)
        print("已保存。")
        n += 1
        # except selenium.common.exceptions.NoSuchWindowException as ex2:
        #     print("爬虫终止")
        #     pass
        # except Exception as ex1:
        #     all_urls.append(url)
        #     save_state(all_urls, crawled_urls)
        #     print("出错啦！！！！！！",ex1)
        #     pass

main()

爬取时的效果图：

爬下来的表白墙：

有没有学到呢？以后就可以看有没有人偷偷给自己写表白墙啦！