javascript - 逐行处理文件，并在输出中保持行顺序

Question

所以我正在编写一个 NodeJS 路由，用户上传一个文件(缓冲区)，逐行处理(需要调用 RESP API 来处理每一行)，然后将结果输出到另一个缓冲区，发送作为下载文件提供给用户。

这是路由代码:

app.post('/tokenizeFile', isLoggedIn, upload.single('file'), function(req, res){
        var file = req.file;

        //File Validations
        if (!validator.validateFile(file)) res.redirect('/?err=invalidFile');

        //Process file
        tokenizer.tokenizeFile(file, req).then((data)=>{
            //res.setHeader('Content-Length', stat.size);
            res.setHeader('Content-Type', 'text/plain');
            res.setHeader('Content-Disposition', 'attachment; filename=tokenized.txt');
            res.write(data, 'binary');
            res.end();
        }).catch((err)=>{
            res.redirect('/?err='+err);
        });

    });

这是 tokenizer.tokenizeFile 代码:

tokenizeFile: function(file, req){

        actionLogger.info(`Request to tokenize ${file.originalname} received. Made by: ${req.user.displayName}`);

        return new Promise(function(resolve, reject){
            var fileProcessPromise = Promise.resolve();

            var lineReader = require('readline').createInterface({
                input: require('streamifier').createReadStream(file.buffer)
            });

            var output = "";

            lineReader.on('line', function (line) {
                //Tokenize each line

                if (!validate.validateLine(line)) return reject(`Invalid line [${line}].`);
                fileProcessPromise = Tokenize(line)
                .then((data)=>{
                    output += data + "\\r\\n";
                })
                .catch((err)=>{
                    reject(`API didn\'t respond.`);
                });
            });

            lineReader.on('close', () => {
                fileProcessPromise.then(()=>resolve(output));
            });

        });

    }

Tokenize 函数返回一个 promise ，因为它是对 RESTful API 的 HTTP 请求。

问题是我需要输出文件保持相同的顺序，使用上面的代码，它的顺序取决于 Tokenize 函数解析的速度。

关于如何实现这一点有什么想法吗？

Answer 1

1)分词文件代码:

tokenizeFile: (file, req) => {

  actionLogger.info(`Request to tokenize ${file.originalname} received. Made by: ${req.user.displayName}`);

  return new Promise((resolve, reject) => {

    const lines = [], responses = [];

    const lineReader = require('readline').createInterface({
      input: require('streamifier').createReadStream(file.buffer)
    });

    // 1. read all lines to array
    lineReader.on('line', line => {
      if(!validate.validateLine(line)) {
        return reject(`Invalid line [${line}].`);
      }
      lines.push(line);
    });


    lineReader.on('close', async () => {
      // 2. process every line sequentially
      try {
        for(const line of lines) {
          const response = await Tokenize(line);
          responses.push(response);
        }
        resolve(responses.join("\n"));
      }
      .catch(error => {
        console.log(error);
        reject("API didn't respond");
      });
    });

  });
}

2) 以及请求部分:

app.post(
  '/tokenizeFile', 
  isLoggedIn, 
  upload.single('file'), 
  async (req, res) => {
    try {
      const file = req.file;

      if (!validator.validateFile(file)) {
        throw new Error('invalidFile');
      }

      const data = await tokenizer.tokenizeFile(file, req);

      res.setHeader('Content-Type', 'text/plain');
      res.setHeader('Content-Disposition', 'attachment; filename=tokenized.txt');
      res.write(data, 'binary');
      res.end();
    })
    .catch(error => {
      res.redirect('/?err='+error);
    });
  });