如何通过 nodejs 在 S3 中创建 Excel 文件？

Question

I'm trying to create a script that will take a JSON object and put it into the S3 bucket as an xlsx file

I am planning on integrating SQL queries to the code but for now limiting it to JSON for easier coding. I've tried using alasql and xlsx to try and create the S3.putObject body but the output creates a corrupted excel file

var data = [{a:1,b:1,c:1},{a:1,b:2,c:1},{a:1,b:3,c:1}, {a:2,b:1,c:1}];
  var a = XLSX.utils.json_to_sheet(data);

  var params = {
  'Body' : Buffer.from(a),
  'Key': event.Key + '.xlsx',
  'Bucket': event.Bucket
};

s3.putObject(params).promise();

I expect that the data will be placed in an xlsx file in the S3 bucket and while the file is created, it is corrupted

Answer 1

What worked for me is this, using sheetjs import xlsx from 'xlsx'; :

        // initiate the workbook
        const wb = xlsx.utils.book_new();

        // add properties to the sheet
        wb.Props = {
          Title: 'Books Borrowed',
          Subject: 'Borrowed Books',
          Author: 'Admin',
          CreatedDate: '2020-04-23',
        };

        // add a sheet
        wb.SheetNames.push('Borrowed');

        // I used aoa_to_sheet because I'm having an issue with json_to_sheet but was able to create a workaround, see: https://github.com/SheetJS/sheetjs/issues/1487#issuecomment-618767784
        // I find the aoa_to_sheet a much cleaner approach
        const ws = xlsx.utils.aoa_to_sheet(sheet);
        wb.Sheets.Borrowed = ws;

        // generate output as buffer
        const wbOut = xlsx.write(wb, {
          bookType: 'xlsx',
          type: 'buffer',
        });

        // upload to S3
        const data = await s3
          .putObject({
            Bucket: config.s3.s3BucketPublic,
            Key: 'filenameHere.xlsx',
            ACL: 'public-read',
            Body: wbOut,
            ContentType:
              'application/vnd.openxmlformats-officedocument.spreadsheetml.sheet',
          })
          .promise();

Answer 2

Below is what works for me.

const Excel = require('exceljs');
const aws = require('aws-sdk');
let workbook = new Excel.Workbook();
workbook.creator = 'System';
workbook.lastModifiedBy = 'System';
workbook.created = new Date();
workbook.modified = new Date();
workbook.lastPrinted = new Date();

let worksheet = workbook.addWorksheet('SOME Data');
worksheet.columns = [{
    header: 'somekey',
    key: 'DeviceID'
}, {
    header: 'Type',
    key: 'Type'
}];

hcpData.forEach(element => {
   worksheet.addRow({
       somekey: element.DEVICEID,
       Type: element.TYPE
   });
});

worksheet.getRow(1).eachCell(cell => {
    cell.font = {
        bold: true
   };
});
//configuring the AWS environment
aws.config.update({
    accessKeyId: 'putyourkye',
    secretAccessKey: 'putyourkye',
});

const s3 = new aws.S3();
const stream = new Stream.PassThrough();

workbook.xlsx.write(stream).then(() => {
    return s3
       .upload({
        Key: Date.now() + '_XYZNAME.xlsx',
        Bucket: 'abc-reports',
        Body: stream,
        ACL: 'public-read',
    })
    .promise()
    .then(data => {
        // do some task after upload
    });
});

Answer 3

This can be achieved by doing the following steps:

1. Install and import ExcelJS import * as excel from "exceljs";

2. Create a function that will generate the excel file as stream:

const createExcel = async (
  headers: Partial<excel.Column>[],
  rows: any[]
): Promise<Buffer> => {
  const workbook: excel.stream.xlsx.WorkbookWriter =
    new excel.stream.xlsx.WorkbookWriter({});
  const sheet: excel.Worksheet = workbook.addWorksheet("My Worksheet");
  sheet.columns = headers;
  for (let i = 0; i < rows.length; i++) {
    sheet.addRow(rows[i]);
  }
  sheet.commit();
  return new Promise((resolve, reject): void => {
    workbook
      .commit()
      .then(() => {
        const stream: any = (workbook as any).stream;
        const result: Buffer = stream.read();
        resolve(result);
      })
      .catch((e) => {
        reject(e);
      });
  });
};

3. Retrieve the data as JSON, in my case from MySQL table: const data = db.query("SELECT * FROM data_table");

4. Generate the excel stream by using above function - createExcel

const stream: Buffer = await createExcel(
      [
        { header: "Column one header", key: "id" },
        { header: "Name column header", key: "name" },
      ],
      data
    );

The first argument will set the excel file column headers. make sure the key will match with the key from your JSON file.

The second argument (data) represents the JSON file which will be an array of objects that have the same properties as the values of each of the keys.

5. Now you can do whatever you need with the stream variable which will contain the Buffer. To upload to S3:

  const BUCKET_NAME = "data-bucket";
  const KEY = `xls-file-name.xlsx`;
  const params = {
    Bucket: BUCKET_NAME,
    Key: KEY,
    Body: stream,
  };

  s3.upload(params, function (err: any, data: any) {
    if (err) {
      console.log(err);
      console.log("Error");
    }
    console.log(`File uploaded successfully. ${data.Location}`);
  });