1.3.3解决报废问题（覆盖image_downloaded方法）：

import scrapy
from scrapy.pipelines.images import ImagesPipeline
from scrapy.utils.misc import md5sum
class MyImagesPipeline(ImagesPipeline):
    def get_media_requests(self, item, info):
        for image_url in item['image_urls']:
            yield scrapy.Request(image_url, meta={'image_names': item["image_names"]})

    def image_downloaded(self, response, request, info):
        checksum = None
        for path, image, buf in self.get_images(response, request, info):
            if checksum is None:
                buf.seek(0)
                checksum = md5sum(buf)
            width, height = image.size
            path = 'full/%s' % response.meta['image_names'][0] #**Here Changed**
            self.store.persist_file(
                path, buf, info,
                meta={'width': width, 'height': height},
                headers={'Content-Type': 'image/jpeg'})
        return checksum

解决方法是重写DownloadImagesPipeline类的image_key方法。

def image_key(self, url):
    return 'image_name.here'

例如，如果需要URL的图像名称，可以使用

url.split('/')[-1]

作为映像的名称。请注意此方法已过时，可以在将来的版本中删除。
或者，您可以在Spider中为图像设置image_name：

item['image_name'] = ['whatever_you_want']

在这种情况下，您必须进一步扩展您的管道，以利用您提供的映像的名称：

def get_media_requests(self, item, info):
        return [Request(x, meta={'image_names': item["image_name"]})
                for x in item.get('image_urls', [])]

def get_images(self, response, request, info):
    for key, image, buf, in super(DownloadImagesPipeline, self).get_images(response, request, info):
        if re.compile('^[0-9,a-f]+.jpg$').match(key):
            key = self.change_filename(key, response)
        yield key, image, buf

def change_filename(self, key, response):
    return "%s.jpg" % response.meta['image_name'][0]

当然，您的管道应该扩展到ImagesPipeline。

它将给予自定义图像名称的答案，以及将此类图像保存到哪个文件夹（自定义命名）。

# spider.py

import scrapy
from ..items import DusharaItem
class DusharaSpider(scrapy.Spider):
    name='dushara'
    start_urls=['https://www.indiaglitz.com/dushara-photos-tamil-actress-3129970-8771']
    def parse(self,response):
        selector = response.xpath('//div[@class="gallmain gallerycontainer-8771"]/div[@class="gallery_detail gal-8771"]')
        for sel in selector:       
            item = DusharaItem()
            item['image_urls']      = sel.xpath('./img/@src').extract_first()
            #item['image_urls']     = [sel.xpath('./img/@src').extract_first()]  # for default scraping process
            item['folder_names_1']  = 'Actress'
            item['folder_names_2']  = 'Tamil'
            item['image_names']     = sel.xpath('./img/@src').extract_first().split('/')[-1] # it should contain image extension like .jpg
            yield item

# items.py

import scrapy
class DusharaItem(scrapy.Item):
    image_urls     = scrapy.Field()
    images         = scrapy.Field()
    folder_names_1 = scrapy.Field()
    folder_names_2 = scrapy.Field()
    image_names    = scrapy.Field()

# pipelines.py

import scrapy
from scrapy.pipelines.images import ImagesPipeline
class DusharaPipeline(ImagesPipeline):
    def get_media_requests(self, item,info):
        url            = item['image_urls']
        folder_names_1 = item['folder_names_1']
        folder_names_2 = item['folder_names_2']
        image_names    = item['image_names']
        yield scrapy.Request(url=url, meta={'folder_names_1': folder_names_1, 'folder_names_2': folder_names_2, 'image_names': image_names})

    def file_path(self, request, response=None, info=None, *, item=None):
        folder_names_1 = request.meta['folder_names_1']
        folder_names_2 = request.meta['folder_names_2']
        image_names    = request.meta['image_names']
        return '/'+folder_names_1+'/'+folder_names_2+'/'+image_names

# settings.py

ITEM_PIPELINES  = {'dushara.pipelines.DusharaPipeline': 300}

# ITEM_PIPELINES = {'scrapy.pipelines.images.ImagesPipeline': 1} # for default scraping process

IMAGES_STORE    = r'D:\Scraped'