|
| 1 | +# Define here the models for your spider middleware |
| 2 | +# |
| 3 | +# See documentation in: |
| 4 | +# https://docs.scrapy.org/en/latest/topics/spider-middleware.html |
| 5 | + |
| 6 | +import time |
| 7 | +from selenium import webdriver |
| 8 | +from scrapy import signals |
| 9 | +from scrapy.http.response.html import HtmlResponse |
| 10 | + |
| 11 | +# useful for handling different item types with a single interface |
| 12 | +from itemadapter import is_item, ItemAdapter |
| 13 | + |
| 14 | + |
| 15 | +class DcdSpiderMiddleware: |
| 16 | + # Not all methods need to be defined. If a method is not defined, |
| 17 | + # scrapy acts as if the spider middleware does not modify the |
| 18 | + # passed objects. |
| 19 | + |
| 20 | + @classmethod |
| 21 | + def from_crawler(cls, crawler): |
| 22 | + # This method is used by Scrapy to create your spiders. |
| 23 | + s = cls() |
| 24 | + crawler.signals.connect(s.spider_opened, signal=signals.spider_opened) |
| 25 | + return s |
| 26 | + |
| 27 | + def process_spider_input(self, response, spider): |
| 28 | + # Called for each response that goes through the spider |
| 29 | + # middleware and into the spider. |
| 30 | + |
| 31 | + # Should return None or raise an exception. |
| 32 | + return None |
| 33 | + |
| 34 | + def process_spider_output(self, response, result, spider): |
| 35 | + # Called with the results returned from the Spider, after |
| 36 | + # it has processed the response. |
| 37 | + |
| 38 | + # Must return an iterable of Request, or item objects. |
| 39 | + for i in result: |
| 40 | + yield i |
| 41 | + |
| 42 | + def process_spider_exception(self, response, exception, spider): |
| 43 | + # Called when a spider or process_spider_input() method |
| 44 | + # (from other spider middleware) raises an exception. |
| 45 | + |
| 46 | + # Should return either None or an iterable of Request or item objects. |
| 47 | + pass |
| 48 | + |
| 49 | + def process_start_requests(self, start_requests, spider): |
| 50 | + # Called with the start requests of the spider, and works |
| 51 | + # similarly to the process_spider_output() method, except |
| 52 | + # that it doesn’t have a response associated. |
| 53 | + |
| 54 | + # Must return only requests (not items). |
| 55 | + for r in start_requests: |
| 56 | + yield r |
| 57 | + |
| 58 | + def spider_opened(self, spider): |
| 59 | + spider.logger.info('Spider opened: %s' % spider.name) |
| 60 | + |
| 61 | + |
| 62 | +class DcdDownloaderMiddleware: |
| 63 | + # Not all methods need to be defined. If a method is not defined, |
| 64 | + # scrapy acts as if the downloader middleware does not modify the |
| 65 | + # passed objects. |
| 66 | + |
| 67 | + def __init__(self): |
| 68 | + # 加载测试浏览器 |
| 69 | + options = webdriver.ChromeOptions() |
| 70 | + options.add_argument('--no-sandbox') |
| 71 | + options.add_argument('--disable-gpu') |
| 72 | + options.add_argument('--ignore-certificate-errors') |
| 73 | + options.add_argument('--ignore-ssl-errors') |
| 74 | + |
| 75 | + self.driver = webdriver.Chrome(executable_path=r"C:\drf2\drf2\chromedriver.exe",options=options) |
| 76 | + self.driver.maximize_window() |
| 77 | + |
| 78 | + @classmethod |
| 79 | + def from_crawler(cls, crawler): |
| 80 | + # This method is used by Scrapy to create your spiders. |
| 81 | + s = cls() |
| 82 | + crawler.signals.connect(s.spider_opened, signal=signals.spider_opened) |
| 83 | + return s |
| 84 | + |
| 85 | + def process_request(self, request, spider): |
| 86 | + print('request.url',request.url) |
| 87 | + self.driver.get(request.url) |
| 88 | + js = 'return document.body.scrollHeight;' |
| 89 | + height = 0 |
| 90 | + if request.url != 'https://www.dongchedi.com/auto/library/x-x-x-x-x-x-x-x-x-x-x-x-x-x-x-x-x-x': |
| 91 | + while True: |
| 92 | + new_height = self.driver.execute_script(js) |
| 93 | + if new_height > height: |
| 94 | + self.driver.execute_script('window.scrollTo(0, document.body.scrollHeight)') |
| 95 | + height = new_height |
| 96 | + time.sleep(1) |
| 97 | + else: |
| 98 | + print("滚动条已经处于页面最下方!") |
| 99 | + break |
| 100 | + source = self.driver.page_source |
| 101 | + # 创建一个response对象,把页面信息都封装在reponse对象中 |
| 102 | + response = HtmlResponse(url=self.driver.current_url,body=source,request = request,encoding="utf-8") |
| 103 | + return response |
| 104 | + |
| 105 | + def process_response(self, request, response, spider): |
| 106 | + # Called with the response returned from the downloader. |
| 107 | + |
| 108 | + # Must either; |
| 109 | + # - return a Response object |
| 110 | + # - return a Request object |
| 111 | + # - or raise IgnoreRequest |
| 112 | + return response |
| 113 | + |
| 114 | + def process_exception(self, request, exception, spider): |
| 115 | + # Called when a download handler or a process_request() |
| 116 | + # (from other downloader middleware) raises an exception. |
| 117 | + |
| 118 | + # Must either: |
| 119 | + # - return None: continue processing this exception |
| 120 | + # - return a Response object: stops process_exception() chain |
| 121 | + # - return a Request object: stops process_exception() chain |
| 122 | + pass |
| 123 | + |
| 124 | + def spider_opened(self, spider): |
| 125 | + spider.logger.info('Spider opened: %s' % spider.name) |
0 commit comments