爬虫 - 爬取王者荣耀所有皮肤图片

结果展示

安装

pip install requests logger

代码

import json
import os
import re
from concurrent.futures import ThreadPoolExecutor

import requests
from loguru import logger


def parse_url(url, b=False):
    try:
        headers = {"User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/86.0.4240.198 Safari/537.36", }
        res = requests.get(url, headers=headers)
        res.encoding = "GBK"
        assert res.status_code == 200, "Code not is 200"
        return res.content if b else res.text
    except:
        pass


def download_img(img_url, hero_name, hero_img, num):
    b_data = parse_url(img_url, b=True)
    if b_data is None:
        return
    with open(hero_img, "wb") as f:
        f.write(b_data)
    logger.success(f"{hero_name} 第{num}张皮肤图片 下载完毕")


def process_hero(id, name):
    logger.info(f"{id}\t{name}\t处理中...")

    hero_dir = f"./英雄皮肤/{name}"
    if not os.path.exists(hero_dir):
        os.makedirs(hero_dir, exist_ok=True)

    with ThreadPoolExecutor(max_workers=20) as pool:
        for num in range(1, 20):
            hero_img = f"{hero_dir}/皮肤_{num}.png"
            if os.path.exists(hero_img):
                logger.warning(f"{hero_img}已下载过，跳过")
                continue
            img_url = f"https://game.gtimg.cn/images/yxzj/img201606/skin/hero-info/{id}/{id}-bigskin-{num}.jpg"
            pool.submit(download_img, img_url, name, hero_img, num)


def start():
    api_url = "https://game.gtimg.cn/images/yxzj/web201706/js/heroid.js"
    text = parse_url(api_url)
    search_result = re.search('var module_exports = ({.*?})', text, re.S)
    hero_info_str = search_result.group(1)
    hero_info_str = re.sub("'", '"', hero_info_str)
    hero_info_dict = json.loads(hero_info_str)

    with ThreadPoolExecutor(max_workers=10) as pool:
        for hero in hero_info_dict:
            name, id = hero_info_dict[hero], hero
            pool.submit(process_hero, id, name)


if __name__ == '__main__':
    start()