有没有对 Scrapy 爬虫很精通的朋友，我目前的爬虫有些问题:

import scrapy
from xxx.items import WorkItem

class XXXSpider(scrapy.Spider):
    name = "xxx"
    allowed_domains = ["example.com"]
    start_urls = ["https://example.com/xx/xx"]

    def parse(self, response):
        year_list = response.xpath('//ul[@class="p-accordion"]/li')
        for year in year_list[:2]:
            release_dates_url_of_year = year.xpath('.//div[@class="genre -s"]/a/@href').extract()
            for date_url in release_dates_url_of_year:
                yield scrapy.Request (
                    url = date_url,
                    callback =self.date_detail_parse
                )

    def date_detail_parse(self, response):
        work_list = response.xpath('.//div[@class="swiper-slide c-low--6"]/div')
        for work in work_list:
            actress_name = work.xpath('.//a[@class="name c-main-font-hover"]/text()').extract_first()
            if actress_name is not None:
                item = WorkItem()
                item['actress_name'] = actress_name
                item['image_hover'] = work.xpath('.//img[@class="c-main-bg lazyload"]/@data-src').extract_first()
                work_detail_url = work.xpath('.//a[@class="img hover"]/@href').extract_first()

                if work_detail_url is not None:
                    yield scrapy.Request (
                        url = work_detail_url,
                        callback = self.work_detail_pares,
                        meta = {'workItem' : item}
                    )

    def work_detail_pares(self, response):
        item = response.meta['workItem']
        pics_list = response.xpath('.//div[@class="swiper-wrapper"]/div')
        pre_images = []  
        for pic in pics_list:
            img_url = pic.xpath('./img/@data-src').extract_first()
            pre_images.append(img_url)
        item['pre_images'] = pre_images
        item['name'] = response.xpath('.//div[@class="p-workPage l-wrap"]/h2/text()').extract_first().strip()
        item['id'] = response.xpath('.//span[@class="c-tag02 c-main-bg-hover c-main-bg"]/../text()').extract_first()
        item['company'] = 'xxx'
        item['release_date'] = response.xpath('.//div[@class="p-workPage__table"]/div[2]//div[@class="item"]/a/text()').extract_first()
        actress_detail_url = response.xpath('.//div[@class="p-workPage__table"]/div[1]//div[@class="item"]/a/@href').extract_first()
        yield scrapy.Request(
            url = actress_detail_url,
            callback = self.actress_detail_pase,
            meta = {'workItem' : item}
        )

    def actress_detail_pase(self, response):
        item = response.meta['workItem']
        item['actress_avatar'] = response.xpath('.//div[@class="swiper-slide"]/img/@data-src').extract_first()
        yield item

这是我的爬虫主程序，year_list我目前就只取了前 2 条数据做测试，发现爬取下来的数据跟网站上相比，很漏爬很多很多数据，而且每次爬取的数量都是同一个，网站上应该有至少几百条数据，而爬虫爬取下来，每次只能爬 35 条数据。而且每次爬取的数据内容都是变化的，这次没爬到的数据可能下一次再启动爬虫他就爬到了，但是总数永远是 35 。

我也尝试过把item包在deepcopy()里，也没有区别。如果不用for循环，就只爬单条数据的话，他又确实可以把完整的数据都爬取下来有没有大佬知道我这问题到底出在哪里？问 GPT 的各种解答方案也试了，还是不行。