scrapy使用selenium3.0

第一步：在spider类的init方法中初始化一个浏览器驱动（webdriver）

如何配置selenium以及配置浏览器驱动在另一篇博客中selenium3.0环境搭建_与市场搏斗-CSDN博客

第二步：在spider类的close方法中关闭浏览器驱动

import os
import time

import scrapy
from selenium import webdriver

from myscrapy.network_log_option import *


class Selensprider(scrapy.Spider):
    name = 'selenspider'
    allow_domains = 'fw.zjjy.xyz'
    start_urls = ['https://www.zjjy.xyz/web/index']

    def __init__(self):
        # 启动代理命令,打开代理浏览器窗口
        cd_dir = r'cd C:\Users\Lenovo\AppData\Local\Google\Chrome\Application'
        start_proxy = r'chrome.exe --remote-debugging-port=9222 --user-data-dir="C:\selenum\AutomationProfile"'
        os.system(cd_dir)
        time.sleep(0.5)
        os.system(start_proxy)
        # 初始浏览器驱动
        options = get_log_options()
        # 配置代理参数
        options.add_experimental_option('debuggerAddress', 'localhost:90222')
        caps = get_caps()
        self.chrome = webdriver.Chrome(options=options, desired_capabilities=caps)
        super().__init__()

    def parse(self, response, **kwargs):
        pass

    # 整个爬虫结束后关闭浏览器
    def close(self, spider):
        self.chrome.quit()

第三步：在中间件中拦截request和response，并对浏览器进行操作

# selenium处理中间件
class SeleniumMiddleware(object):
    def process_response(self, request, response, spider):
        chrome = spider.chrome
        chrome.get(request.url)
        #这里可以对浏览器进行操作，比如浏览器全屏，下拉上拉等加载事件的操作
        chrome.maximize_window()
        # 阻塞5秒，等待ajax加载完毕
        time.sleep(5)
        # network记录，这里可以获得所有ajax请求的结果
        logs = get_xhr_logs(chrome)
        # 动态加载后的网页
        html = chrome.page_source
        return scrapy.http.HtmlResponse(url=request.url, body=html.encode('utf-8'), encoding='utf-8',request=request)

util工具

import json
from selenium import webdriver
from selenium.webdriver import DesiredCapabilities


def get_xhr_logs(chrome):
    log_xhr_array = []
    for typelog in chrome.log_types:
        perfs = chrome.get_log(typelog)
        for row in perfs:
            log_data = row
            message_ = log_data['message']
            try:
                log_json = json.loads(message_)
                log = log_json['message']
                if log['method'] == 'Network.responseReceived':
                    # 去掉静态js、css等，仅保留xhr请求
                    type_ = log['params']['type']
                    if type_ == "XHR":
                        log_xhr_array.append(log)
            except:
                pass
    return log_xhr_array


def get_log_options():
    option = webdriver.ChromeOptions()
    option.add_argument('--no-sandbox')
    option.add_argument('--headless')
    option.add_argument("--disable-extensions")
    option.add_argument("--allow-running-insecure-content")
    option.add_argument("--ignore-certificate-errors")
    option.add_argument("--disable-single-click-autofill")
    option.add_argument("--disable-autofill-keyboard-accessory-view[8]")
    option.add_argument("--disable-full-form-autofill-ios")
    option.add_argument('user-agent=Mozilla/5.0 (Macintosh; Intel Mac OS X 10.14; rv:55.0) Gecko/20100101 Firefox/55.0')
    option.add_experimental_option('w3c', False)
    option.add_experimental_option('perfLoggingPrefs', {
        'enableNetwork': True,
        'enablePage': False,
    })
    return option


def get_caps():
    caps = DesiredCapabilities.CHROME
    caps['loggingPrefs'] = {
        'browser': 'ALL',
        'performance': 'ALL',
    }
    caps['perfLoggingPrefs'] = {
        'enableNetwork': True,
        'enablePage': False,
        'enableTimeline': False
    }
    return caps

第四步：在setting中启用此中间件。

DOWNLOADER_MIDDLEWARES = {
   'asong.middlewares.SeleniumMiddleware': 543,
}