记录一次线上数据图源本地化操作的过程

最近学了一个比较赞的电商项目，项目作者提供了完整的示例数据，包括商品信息及配图，但是这些配图是固定的URL，商品详情为html，html中有img标签，img标签中也有url。根据过往经验这种在线CDN很容易挂掉，因此产生了把商品数据中的商品图片提取出来，放在自己的腾讯云服务器中的想法，保证可访问性。

演示数据

[{
    "ID": "b93e59e214fc4478ac72652a2c87fe54",
    "GOODS_SERIAL_NUMBER": "2300000059885",
    "SHOP_ID": "402880e860166f3c0160167897d60002",
    "SUB_ID": "402880e86016d1b5016016dcd7c50004",
    "GOOD_TYPE": 1,
    "STATE": 0,
    "IS_DELETE": 1,
    "NAME": "云南红提800g/盒",
    "ORI_PRICE": 18,
    "PRESENT_PRICE": 15,
    "AMOUNT": 10000,
    "DETAIL": "<img src=\"http://images.koow.cc/shopGoodsDetailImg/20171225/20171225112029_9395.jpg\" width=\"100%\" height=\"auto\" alt=\"\" /><img src=\"http://images.koow.cc/shopGoodsDetailImg/20171225/20171225112029_3391.jpg\" width=\"100%\" height=\"auto\" alt=\"\" /><img src=\"http://images.koow.cc/shopGoodsDetailImg/20171225/20171225112029_7603.jpg\" width=\"100%\" height=\"auto\" alt=\"\" /><img src=\"http://images.koow.cc/shopGoodsDetailImg/20171225/20171225112029_4718.jpg\" width=\"100%\" height=\"auto\" alt=\"\" /><img src=\"http://images.koow.cc/shopGoodsDetailImg/20171225/20171225112030_778.jpg\" width=\"100%\" height=\"auto\" alt=\"\" /><img src=\"http://images.koow.cc/shopGoodsDetailImg/20171225/20171225112030_2602.jpg\" width=\"100%\" height=\"auto\" alt=\"\" /><img src=\"http://images.koow.cc/shopGoodsDetailImg/20171225/20171225112030_7913.jpg\" width=\"100%\" height=\"auto\" alt=\"\" /><img src=\"http://images.koow.cc/shopGoodsDetailImg/20171225/20171225112030_202.jpg\" width=\"100%\" height=\"auto\" alt=\"\" /><img src=\"http://images.koow.cc/shopGoodsDetailImg/20171225/20171225112030_4296.jpg\" width=\"100%\" height=\"auto\" alt=\"\" /><img src=\"http://images.koow.cc/shopGoodsDetailImg/20171225/20171225112030_6956.jpg\" width=\"100%\" height=\"auto\" alt=\"\" /><img src=\"http://images.koow.cc/shopGoodsDetailImg/20171225/20171225112030_8200.jpg\" width=\"100%\" height=\"auto\" alt=\"\" /><img src=\"http://images.koow.cc/shopGoodsDetailImg/20171225/20171225112031_3967.jpg\" width=\"100%\" height=\"auto\" alt=\"\" /><img src=\"http://images.koow.cc/shopGoodsDetailImg/20171225/20171225112031_5114.jpg\" width=\"100%\" height=\"auto\" alt=\"\" />",
    "BRIEF": null,
    "SALES_COUNT": 0,
    "IMAGE1": "http://images.koow.cc/shopGoodsImg/20171225/20171225112020_561.jpg",
    "IMAGE2": null,
    "IMAGE3": null,
    "IMAGE4": null,
    "IMAGE5": null,
    "ORIGIN_PLACE": null,
    "GOOD_SCENT": null,
    "CREATE_TIME": 1514172047397,
    "UPDATE_TIME": 1522037064430,
    "IS_RECOMMEND": 0,
    "PICTURE_COMPERSS_PATH": "http://images.koow.cc/compressedPic/20171225112020_561.jpg"
  },
  {
    "ID": "e0ab2f6e2802443ba117b1146cf85fee",
    "GOODS_SERIAL_NUMBER": "4894375014863",
    "SHOP_ID": "402880e860166f3c0160167897d60002",
    "SUB_ID": "2c9f6c94609a62be0160a02d1dc20021",
    "GOOD_TYPE": 1,
    "STATE": 0,
    "IS_DELETE": 1,
    "NAME": "菓子町园道乳酸菌味夹心饼干（抹茶味）540/罐",
    "ORI_PRICE": 29.8,
    "PRESENT_PRICE": 29.8,
    "AMOUNT": 10000,
    "DETAIL": "<img src=\"http://images.koow.cc/shopGoodsDetailImg/20180213/20180213110655_230.jpg\" width=\"100%\" height=\"auto\" alt=\"\" /><img src=\"http://images.koow.cc/shopGoodsDetailImg/20180213/20180213110656_329.jpg\" width=\"100%\" height=\"auto\" alt=\"\" /><img src=\"http://images.koow.cc/shopGoodsDetailImg/20180213/20180213110656_2659.jpg\" width=\"100%\" height=\"auto\" alt=\"\" /><img src=\"http://images.koow.cc/shopGoodsDetailImg/20180213/20180213110656_9521.jpg\" width=\"100%\" height=\"auto\" alt=\"\" /><img src=\"http://images.koow.cc/shopGoodsDetailImg/20180213/20180213110656_8611.jpg\" width=\"100%\" height=\"auto\" alt=\"\" /><img src=\"http://images.koow.cc/shopGoodsDetailImg/20180213/20180213110656_1390.jpg\" width=\"100%\" height=\"auto\" alt=\"\" /><img src=\"http://images.koow.cc/shopGoodsDetailImg/20180213/20180213110656_7291.jpg\" width=\"100%\" height=\"auto\" alt=\"\" /><img src=\"http://images.koow.cc/shopGoodsDetailImg/20180213/20180213110657_3919.jpg\" width=\"100%\" height=\"auto\" alt=\"\" /><img src=\"http://images.koow.cc/shopGoodsDetailImg/20180213/20180213110657_2170.jpg\" width=\"100%\" height=\"auto\" alt=\"\" /><img src=\"http://images.koow.cc/shopGoodsDetailImg/20180213/20180213110657_4402.jpg\" width=\"100%\" height=\"auto\" alt=\"\" /><img src=\"http://images.koow.cc/shopGoodsDetailImg/20180213/20180213110657_1926.jpg\" width=\"100%\" height=\"auto\" alt=\"\" /><img src=\"http://images.koow.cc/shopGoodsDetailImg/20180213/20180213110657_9438.jpg\" width=\"100%\" height=\"auto\" alt=\"\" /><img src=\"http://images.koow.cc/shopGoodsDetailImg/20180213/20180213110657_4361.jpg\" width=\"100%\" height=\"auto\" alt=\"\" /><img src=\"http://images.koow.cc/shopGoodsDetailImg/20180213/20180213110657_2730.jpg\" width=\"100%\" height=\"auto\" alt=\"\" /><img src=\"http://images.koow.cc/shopGoodsDetailImg/20180213/20180213110658_314.jpg\" width=\"100%\" height=\"auto\" alt=\"\" /><img src=\"http://images.koow.cc/shopGoodsDetailImg/20180213/20180213110658_8779.jpg\" width=\"100%\" height=\"auto\" alt=\"\" /><img src=\"http://images.koow.cc/shopGoodsDetailImg/20180213/20180213110658_9878.jpg\" width=\"100%\" height=\"auto\" alt=\"\" /><img src=\"http://images.koow.cc/shopGoodsDetailImg/20180213/20180213110658_3471.jpg\" width=\"100%\" height=\"auto\" alt=\"\" />",
    "BRIEF": null,
    "SALES_COUNT": 0,
    "IMAGE1": "http://images.koow.cc/shopGoodsImg/20180213/20180213110648_2744.jpg",
    "IMAGE2": null,
    "IMAGE3": null,
    "IMAGE4": null,
    "IMAGE5": null,
    "ORIGIN_PLACE": null,
    "GOOD_SCENT": null,
    "CREATE_TIME": 1518491222336,
    "UPDATE_TIME": 1523174099461,
    "IS_RECOMMEND": 0,
    "PICTURE_COMPERSS_PATH": "http://images.koow.cc/compressedPic/20180213110648_2744.jpg"
  }]

可以看到，数据比较完整，包括ID、编号、名称、价格、介绍等信息。
如果想要提取JSON对象中的图片URL，对于其中的images1-images5对象比较好处理，只需要遍历即可。对于DETAIL中的图片URL，由于URL混在html中，没有办法直接拿到，可通过正则匹配的形式获取。下面分步骤操作：

提取IMAGE1-IMAGE5中的图片URL

const fs = require("fs");

fs.readFile("./goods_demo.json", "utf8", (err, data) => {
  // 序列化数据 
  data = JSON.parse(data);
  data.map((value, index) => {
    for (let i = 0; i < 5; i++) {
      // 遍历数据，并写入到名为result.txt的文件中
      if (value[`IMAGE${i + 1}`] !== null) {
        const url = value[`IMAGE${i + 1}`]
        fs.appendFile("./result.txt",`\r\n${url}`, function(err) {
          if (err) console.log("写文件操作失败");
          else console.log("写文件操作成功");
        });
      }
    }
  });
});

使用NodeJS运行上面的代码后，就能够正确的读取到IMAGE对象中的URL,并写入到result.txt文件中。

提取DETAIL对象中的图片URL

对url地址分析可以发现，图片URL包括http开头(part1)，CDN的URL(part2)，图片所在的目录(part3)，图片的名称(part4)：

"http://(part1)images.koow.cc(part2)/shopGoodsImg(part3)/20171225(part3)/20171225112020_561.jpg(part4)"

根据以上正则规则，可以用以下正则进行匹配！

// \w表示任意字母数字或下划线
// url中的/符号需要转义
// {2,5}表示出现2-5次
// /g表示全局匹配
const urlReg = /http\:\/\/images.koow.cc(\/\w+){2,5}\.jpg/g;

加上对JSON中DETAIL对象处理的代码以后，整体代码如下：

const fs = require("fs");

fs.readFile("./goods_demo.json", "utf8", (err, data) => {
  data = JSON.parse(data);
  data.map((value, index) => {
    if (value.DETAIL) {
      // 匹配图片的正则表达式
      const urlReg = /http\:\/\/images.koow.cc(\/\w+){2,5}\.jpg/g;
      const arrlist = value.DETAIL.match(urlReg);
      // 对匹配到的image list遍历并写入文件
      if (arrlist && arrlist.length) {
        arrlist.map(item => {
          fs.appendFile("./result.txt", `\r\n${item}`, function(err) {
            if (err) console.log("写DETAIL记录操作失败");
            else console.log("写DETAIL记录操作成功");
          });
        });
      }
    }
    
    for (let i = 0; i < 5; i++) {
      if (value[`IMAGE${i + 1}`] !== null) {
        const url = value[`IMAGE${i + 1}`]
        fs.appendFile("./result.txt",`\r\n${url}`, function(err) {
          if (err) console.log("写文件操作失败");
          else console.log("写文件操作成功");
        });
      }
    }
  });
});

最终提取的url在reuslt.txt中存储，等待后续的处理。

http://images.koow.cc/shopGoodsDetailImg/20171225/20171225112029_9395.jpg
http://images.koow.cc/shopGoodsDetailImg/20171225/20171225112029_3391.jpg
http://images.koow.cc/shopGoodsDetailImg/20171225/20171225112029_4718.jpg
http://images.koow.cc/shopGoodsDetailImg/20171225/20171225112029_7603.jpg
……

批量下载

想要做私有的CDN服务器，文件的存储路径是不能变的，不然就匹配不到数据库中存储的路径。如何在批量下载时保持图片的目录不变呢？很简单，只需要使用wget命令：

wget -nc -r -i ./result.txt

-nc, --no-clobber 不要覆盖已经存在的文件
-r, –recursive 递归下载，下载所有文件
-i, --input-file 下载指定文件中的URL

总结

对JSON或XML数据执行处理是程序员的必备技能，掌握高效的数据处理方法能让工作事半功倍，避免不必要的时间开销。作者写本文的目的是希望能帮助到有同样需求的小伙伴，也希望电脑旁的你能把自己处理数据的技巧分享出来！