使用Node解析XLSX并创建json

Question

好的，所以我发现这个真正有据可查的node_module称为js-xlsx

问题：如何解析xlsx以输出json？

以下是Excel工作表的样子：

最后，json应如下所示：

[
   {
   "id": 1,
   "Headline": "Team: Sally Pearson",
   "Location": "Austrailia",
   "BodyText": "...",
   "Media: "..."
   },
   {
   "id": 2,
   "Headline": "Team: Rebeca Andrade",
   "Location": "Brazil",
   "BodyText": "...",
   "Media: "..."
   }
]

index.js：

if(typeof require !== 'undefined') {
    console.log('hey');
    XLSX = require('xlsx');
}
var workbook = XLSX.readFile('./assets/visa.xlsx');
var sheet_name_list = workbook.SheetNames;
sheet_name_list.forEach(function(y) { /* iterate through sheets */
  var worksheet = workbook.Sheets[y];
  for (z in worksheet) {
    /* all keys that do not begin with "!" correspond to cell addresses */
    if(z[0] === '!') continue;
    // console.log(y + "!" + z + "=" + JSON.stringify(worksheet[z].v));

  }

});
XLSX.writeFile(workbook, 'out.xlsx');

Answer 1

改进版的“Josh Marinacci”答案，它将超越Z列（即AA1）。

var XLSX = require('xlsx');
var workbook = XLSX.readFile('test.xlsx');
var sheet_name_list = workbook.SheetNames;
sheet_name_list.forEach(function(y) {
    var worksheet = workbook.Sheets[y];
    var headers = {};
    var data = [];
    for(z in worksheet) {
        if(z[0] === '!') continue;
        //parse out the column, row, and value
        var tt = 0;
        for (var i = 0; i < z.length; i++) {
            if (!isNaN(z[i])) {
                tt = i;
                break;
            }
        };
        var col = z.substring(0,tt);
        var row = parseInt(z.substring(tt));
        var value = worksheet[z].v;

        //store header names
        if(row == 1 && value) {
            headers[col] = value;
            continue;
        }

        if(!data[row]) data[row]={};
        data[row][headers[col]] = value;
    }
    //drop those first two rows which are empty
    data.shift();
    data.shift();
    console.log(data);
});

Answer 2

你也可以使用

var XLSX = require('xlsx');
var workbook = XLSX.readFile('Master.xlsx');
var sheet_name_list = workbook.SheetNames;
console.log(XLSX.utils.sheet_to_json(workbook.Sheets[sheet_name_list[0]]))

Answer 3

我认为这段代码会做你想要的。它将第一行存储为一组标头，然后将其余存储在一个数据对象中，您可以将其作为JSON写入磁盘。

var XLSX = require('xlsx');
var workbook = XLSX.readFile('test.xlsx');
var sheet_name_list = workbook.SheetNames;
sheet_name_list.forEach(function(y) {
    var worksheet = workbook.Sheets[y];
    var headers = {};
    var data = [];
    for(z in worksheet) {
        if(z[0] === '!') continue;
        //parse out the column, row, and value
        var col = z.substring(0,1);
        var row = parseInt(z.substring(1));
        var value = worksheet[z].v;

        //store header names
        if(row == 1) {
            headers[col] = value;
            continue;
        }

        if(!data[row]) data[row]={};
        data[row][headers[col]] = value;
    }
    //drop those first two rows which are empty
    data.shift();
    data.shift();
    console.log(data);
});

打印出来

[ { id: 1,
    headline: 'team: sally pearson',
    location: 'Australia',
    'body text': 'majority have…',
    media: 'http://www.youtube.com/foo' },
  { id: 2,
    headline: 'Team: rebecca',
    location: 'Brazil',
    'body text': 'it is a long established…',
    media: 'http://s2.image.foo/' } ]

Answer 4

这里是角度5方法版本，对于那些与y，z，tt在接受的答案中挣扎的人，没有明确的语法。用法：parseXlsx().subscribe((data)=> {...})

parseXlsx() {
    let self = this;
    return Observable.create(observer => {
        this.http.get('./assets/input.xlsx', { responseType: 'arraybuffer' }).subscribe((data: ArrayBuffer) => {
            const XLSX = require('xlsx');
            let file = new Uint8Array(data);
            let workbook = XLSX.read(file, { type: 'array' });
            let sheetNamesList = workbook.SheetNames;

            let allLists = {};
            sheetNamesList.forEach(function (sheetName) {
                let worksheet = workbook.Sheets[sheetName];
                let currentWorksheetHeaders: object = {};
                let data: Array<any> = [];
                for (let cellName in worksheet) {//cellNames example: !ref,!margins,A1,B1,C1

                    //skipping serviceCells !margins,!ref
                    if (cellName[0] === '!') {
                        continue
                    };

                    //parse colName, rowNumber, and getting cellValue
                    let numberPosition = self.getCellNumberPosition(cellName);
                    let colName = cellName.substring(0, numberPosition);
                    let rowNumber = parseInt(cellName.substring(numberPosition));
                    let cellValue = worksheet[cellName].w;// .w is XLSX property of parsed worksheet

                    //treating '-' cells as empty on Spot Indices worksheet
                    if (cellValue.trim() == "-") {
                        continue;
                    }

                    //storing header column names
                    if (rowNumber == 1 && cellValue) {
                        currentWorksheetHeaders[colName] = typeof (cellValue) == "string" ? cellValue.toCamelCase() : cellValue;
                        continue;
                    }

                    //creating empty object placeholder to store current row
                    if (!data[rowNumber]) {
                        data[rowNumber] = {}
                    };

                    //if header is date - for spot indices headers are dates
                    data[rowNumber][currentWorksheetHeaders[colName]] = cellValue;

                }

                //dropping first two empty rows
                data.shift();
                data.shift();
                allLists[sheetName.toCamelCase()] = data;
            });

            this.parsed = allLists;

            observer.next(allLists);
            observer.complete();
        })
    });
}

Answer 5

我发现了一种更好的方法

  function genrateJSONEngine() {
    var XLSX = require('xlsx');
    var workbook = XLSX.readFile('test.xlsx');
    var sheet_name_list = workbook.SheetNames;
    sheet_name_list.forEach(function (y) {
      var array = workbook.Sheets[y];

      var first = array[0].join()
      var headers = first.split(',');

      var jsonData = [];
      for (var i = 1, length = array.length; i < length; i++) {

        var myRow = array[i].join();
        var row = myRow.split(',');

        var data = {};
        for (var x = 0; x < row.length; x++) {
          data[headers[x]] = row[x];
        }
        jsonData.push(data);

      }