Python - 爬虫 使用scrapy框架获取豆瓣图片信息

 

 

下面是主代码逻辑，非常简单

# coding:utf-8

import json
import scrapy
from ..items import DouyuItem


class DouyuSpider(scrapy.Spider):
    name = "douyu"
    allowed_domains = ["douyucdn.cn"]

    base_url = "http://capi.douyucdn.cn/api/v1/getVerticalRoom?limit=100&offset="
    offset = 0

    start_urls = [base_url + str(offset)]

    def parse(self, response):
        data_list = json.loads(response.body)['data']

        # 判断响应数据是否为空，如果为空直接退出
        if not data_list:
            return

        for data in data_list:
            item = DouyuItem()

            item['room_link'] = u"http://www.douyu.com/" + data['room_id']
            item['image_src'] = data['vertical_src']
            item['nick_name'] = data['nickname']
            item['anchor_city'] = data['anchor_city']

            # 发送图片的链接请求，传递nick_name字符串并调用回调函数parse_page处理响应
            yield scrapy.Request(item['image_src'], meta={"name": item['nick_name']}, callback=self.parse_image)

        self.offset += 100
        yield scrapy.Request(self.base_url + str(self.offset), callback=self.parse)

    # 1. 通用方式，直接获取资源文件的链接，发送请求再用open方法写入响应数据保存
    def parse_image(self, response):
        filename = response.meta['name'] + ".jpg"

        # 获取响应，按 "wb" 模式保存图片数据
        with open(filename, "wb") as f:
            f.write(response.body)

 items

# -*- coding: utf-8 -*-

import scrapy


class DouyuItem(scrapy.Item):
    # 直播间链接
    room_link = scrapy.Field()
    # 图片链接
    image_src = scrapy.Field()
    # 艺名
    nick_name = scrapy.Field()
    # 所在城市
    anchor_city = scrapy.Field()
    # 图片在磁盘中的路径
    image_path = scrapy.Field()

 管道

# -*- coding: utf-8 -*-

import os
import logging
import scrapy
from scrapy.pipelines.images import ImagesPipeline
from .settings import IMAGES_STORE


class DouyuImagesPipeline(ImagesPipeline):
    def get_media_requests(self, item, info):
        # 发送每个图片的请求，响应数据自动保存在 settings.py 的 IMAGES_STORE字段对应的路径里
        yield scrapy.Request(item['image_src'])

    def item_completed(self, results, item, info):

        old_name = IMAGES_STORE + [x['path'] for ok, x in results if ok][0]
        new_name = IMAGES_STORE + item['nick_name'] + ".jpg"

        try:
            os.rename(old_name, new_name)
        except Exception as e:
            logging.error(e)

        return item



class DouyuPipeline(object):
    def process_item(self, item, spider):
        return item

 运行结果