javascript 简单的脚本，从可检测图像的文件名中删除尾随的“垃圾”

Posted 2021-05-03

const fs = require('fs');
const readdir = require('recursive-readdir');

/*
  Simple script that removes trailing "garbage" from detectable images' filenames (i.e. downloads via `wget -i`)
  Eg. image1.jpg.123[1] ==> image1.jpg
*/

const dir = process.argv[2];

if (!dir) {
  console.log('Please specify a directory to process - Aborting.');
  return;
}

const img1 = /^.*\.(jpe?g|png|gif)$/i; // does it end with an image tag?
const img2 = /\.(jpe?g|png|gif)(.*)/i; // group the image tag with other parts

function renameFile(fileName) {
  const matched = fileName.match(img2);
  if (!matched) return;

  const type = matched[1];
  const cruft = matched[2];

  let i = fileName.indexOf(cruft);
  let goodName = fileName.substr(0, i);

  // hack, just make sure it's a unique filename
  i = goodName.indexOf(type);
  goodName = fileName.substr(0, i) + '_.' + type; // eg. 1_.jpg

  fs.renameSync(fileName, goodName);
  console.log(fileName, ' --> ', goodName);
}

(async () => {
  const fileNames = await readdir(dir);
  fileNames.forEach(f => {
    if (!img1.test(f)) renameFile(f);
  });
})();