naver_scraper_spider.py 文件源码

def parse_download_image(self, response):
        # url = "http://postfiles1.naver.net/20141204_160/kdk926_14176651128371lr8c_JPEG/20141203_110254.jpg?type=w2"
        # file_name = 'D:/workspace/DjangoProjects/BlogWorkspace/aquam/media/images/abcd.jpg'
        download_local_url = settings.BASE_DIR + '/media/images/'    #mac test
        # download_local_url = settings.MEDIA_ROOT + '/images/'  #ubuntu
        replace_item = response
        for i in range(0, response.count('src="')):
            temp = response.split('src="')[i+1]
            url = temp.split('"')[0]
            file_name = url.split('/')[-1].split('?')[0].replace('%', '')
            download_url = download_local_url + file_name
            media_url = settings.MEDIA_URL + 'images/' + file_name
            replace_item = str(replace_item).replace(url, media_url)
            if ImageItem.django_model.objects.filter(file='images/' + file_name).count() == 0:    #??? ????
                urllib.request.urlretrieve(url, download_url)      #??? ????
        return replace_item