scrapy从入门到放弃学习项目4

news/2025/2/26 5:38:21

scrapy框架基于selenium，多页面爬取简书文章内容、作者，ajax技术传递的数据等

保存到MongoDB中

class JieshuxSpider(CrawlSpider):
    name = 'jieshux'
    allowed_domains = ['jianshu.com']
    start_urls = ['https://www.jianshu.com/']

    rules = (
        Rule(LinkExtractor(allow=r'.*/p/[a-z0-9]{12}.*'), callback='parse_item', follow=True),
    )

    def parse_item(self, response):
        title = response.xpath('//h1[@class="title"]/text()').get()
        content = response.xpath('//div[@class="show-content-free"]').get()
        author = response.xpath('//div[@class="info"]/span[@class="name"]/a/text()').get()
        publish_time = response.xpath('//div[@class="info"]/div/span[@class="publish-time"]/text()').get()\
            .replace('*', '')
        word_count = response.xpath('//div[@class="info"]/div/span[@class="wordage"]/text()').get().split(' ')[-1]
        read_count = response.xpath('//div[@class="info"]/div/span[@class="views-count"]/text()').get().split(' ')[-1]
        comment_count = response.xpath('//div[@class="info"]/div/span[@class="comments-count"]/text()').get().split(' ')[-1]
        like_count = response.xpath('//div[@class="info"]/div/span[@class="likes-count"]/text()').get().split(' ')[-1]
        subprocess = response.xpath('//div[@class="include-collection"]').get()
        art_id = response.url.split('?')[0].split('/p/')[-1]
        items = JieshuItem(title=title,
                           content=content,
                           art_id=art_id,
                           author=author,
                           publish_time=publish_time,
                           url=response.url,
                           word_count=word_count,
                           read_count=read_count,
                           comment_count=comment_count,
                           like_count=like_count,
                           subprocess=subprocess)
        yield items

jieshuspider

class JieshuSeleniumDownloaderMiddleware(object):

    def __init__(self):
        self.chromedriver = webdriver.Chrome(executable_path=CHROMEDRIVER_PATH)

    def process_request(self, request, spider):
        self.chromedriver.get(request.url)
        time.sleep(1)
        while 1:
            try:
                more = self.chromedriver.find_element_by_class_name('show_more')
                more.click()
                time.sleep(0.3)
            except:
                break
        source = self.chromedriver.page_source.encode('utf-8')
        response = HtmlResponse(url=self.chromedriver.current_url, body=source, request=request)
        return response

middleware

class JieshuPipeline(object):
    def __init__(self):
        self.db = pymongo.MongoClient('127.0.0.1:27017')

    def open_spider(self, spider):

        self.database = self.db['jianshu']
        self.table = self.database['jianshu']

    def process_item(self, item, spider):
        self.table.insert_one(dict(item))
        return item

    def close_spider(self, spider):
        self.db.close()

pipeline

class JieshuItem(scrapy.Item):
    # define the fields for your item here like:
    # name = scrapy.Field()
    title = scrapy.Field()
    content = scrapy.Field()
    art_id = scrapy.Field()
    author = scrapy.Field()
    publish_time = scrapy.Field()
    url = scrapy.Field()
    word_count = scrapy.Field()
    read_count = scrapy.Field()
    like_count = scrapy.Field()
    comment_count = scrapy.Field()
    subprocess = scrapy.Field()