scrapy爬虫系列之三--爬取图片保存到本地及日志的基本用法

import scrapy
import json
from Douyu.items import DouyuItem

class DouyuSpider(scrapy.Spider):
    name = ‘douyu‘
    allowed_domains = [‘douyucdn.cn‘]
    base_url = "http://capi.douyucdn.cn/api/v1/getVerticalRoom?limit=20&offset={}"
    offset = 0
    start_urls = [base_url.format(offset)]

    # 只能返回Request、BaseItem、dict or None
    def parse(self, response):
        data_list = json.loads(response.body)["data"]
        # 如果没有读到数据，则退出，否则继续请求
        if len(data_list) == 0:
            return

        for data in data_list:
            item = DouyuItem()
            item["nickname"] = data["nickname"]
            item["image_url"] = data["vertical_src"]
            yield item

        self.offset += 20
        yield scrapy.Request(self.base_url.format(self.offset), callback=self.parse)

from scrapy.pipelines.images import ImagesPipeline
import scrapy
import os
from Douyu.settings import IMAGES_STORE as images_store      # 读取配置文件的信息


# 继承ImagePipeline，实现对图片的操作，要进行settings的设置：IMAGES_STORE（图片保存位置）、USER_AGENT、ROBOTSTXT_OBEY、ITEM_PIPELINES
class DouyuPipeline(ImagesPipeline):
    # 重写方法
    def get_media_requests(self, item, info):
        image_url = item["image_url"]
        yield scrapy.Request(image_url)

    # 保存图片时重命名
    def item_completed(self, results, item, info):
        # print(results)
        # print("*"* 30)
        # 列表推导式，获取图片的保存路径
        image_url = [x["path"] for ok, x in results if ok]

        # 重命名，由于都是jpg文件，所以直接拼上了
        os.rename(images_store+image_url[0], images_store + item["nickname"] + ".jpg")