从S3下载xlsx并解析

Question

我需要一项服务来从Amazon S3下载excel文件，然后使用node-xlsx进行解析

问题是我无法让xlsx解析文件。当我尝试回读刚写的文件时，代码中找不到该文件。

我不确定这是否是最好的方法，但这是我到目前为止得到的：

router.get('/process', (req, res) => {
    var fileName = 'https://some-bucket.s3.amazonaws.com/some-excel-file.xlsx'
    https.get(fileName, response => {
        var body = ''
        response.on('data', chunk => body += chunk)
        response.on('end', () => {

            //fs is being imported early on this file
            fs.writeFile(__dirname + '/test.xlsx', body)

            var f = fs.createReadStream(__dirname + '/test.xlsx')

            var book = xlsx.parse(f)
            book.forEach(sheet => console.log('sheet', sheet.name) )

            res.status(200)          
        })
        .on('error', e => {
            res.status(500)
        })
    })
    return
})

Answer 1

这样就可以从S3 nodejs读取文件并将其保存在内存中，而无需先将文件写入磁盘上的某个位置。它可以与S3和AWS Lambda结合使用，因此您不必将文件写入Lambda上的某个位置。

请记住，此过程是异步的。

   var params = {
        Bucket: "",
        Key: ""
    };

    var file = s3.getObject(params).createReadStream();
    var buffers = [];

    file.on('data', function (data) {
        buffers.push(data);
    });

    file.on('end', function () {
        var buffer = Buffer.concat(buffers);
        var workbook = xlsx.parse(buffer);
        console.log("workbook", workbook)
    });

Answer 2

如果你想使用异步/等待，这里的解决方案：

    const AWS = require('aws-sdk');
    const XLSX = require('xlsx');

    AWS.config.update({
      accessKeyId: AMAZON_ACCESS_KEY,
      secretAccessKey: AMAZON_SECRET_ACCESS_KEY,
    });

    // Get buffered file from s3    
    function getBufferFromS3(file, callback){
      const buffers = [];
      const s3 = new AWS.S3();
      const stream = s3.getObject({ Bucket: 'yor_buket', Key: file}).createReadStream();
      stream.on('data', data => buffers.push(data));
      stream.on('end', () => callback(null, Buffer.concat(buffers)));
      stream.on('error', error => callback(error));
    }

    // promisify read stream from s3
    function getBufferFromS3Promise(file) {
      return new Promise((resolve, reject) => {
        getBufferFromS3(file, (error, s3buffer) => {
          if (error) return reject(error);
          return resolve(s3buffer);
        });
      });
    };

    // create workbook from buffer
    const buffer = await getBufferFromS3Promise(file);
    const workbook = XLSX.read(buffer);


    // If you want to send the workbook as a download to the api end point in node
    const fileName = "Categories.xlsx";
    res.setHeader('Content-disposition', 'attachment; filename=' + fileName);
    res.setHeader('Content-type', 'application/vnd.openxmlformats-officedocument.spreadsheetml.sheet');
    const wbout = XLSX.write(workbook, { bookType: 'xlsx', type: 'buffer'});
    res.send(new Buffer(wbout));