node-excel-stream

A utility to read and write excels using streams in Node

This package provides ExcelWriter and ExcelReader classes that can be used to writer and parse excel respectively

Installation

npm install node-excel-stream

Usage with es5

// with es5
var ExcelReader = require('node-excel-stream').ExcelReader;
var ExcelWriter = require('node-excel-stream').ExcelWriter;
// or with es6 imports
import { ExcelReader, ExcelWriter } from 'node-excel-stream';

Usage

Parsing Workbooks

Workbooks must be given a schema before they are parsed. If the excel format does not match the given schema, then an error is thrown. Then, each row in each sheet can be parsed using the eachRow function, which returns a promise when resolves after all rows are parsed.

Input: data.xlsx, Sheet: Users

User Name	Value
John	10
Rohan	30
Pooja	50

Parsing:

let dataStream = fs.createReadStream('data.xlsx');
let reader = new ExcelReader(dataStream, {
    sheets: [{
        name: 'Users',
        rows: {
            headerRow: 1,
            allowedHeaders: [{
                name: 'User Name',
                key: 'userName'
            }, {
                name: 'Value',
                key: 'value',
                type: Number
            }]
        }
    }]
})
console.log('starting parse');
reader.eachRow((rowData, rowNum, sheetSchema) => {
    console.log(rowData);
})
.then(() => {
    console.log('done parsing');
});

Output:

starting parse
{ userName: 'John', value: 10 }
{ userName: 'Rohan', value: 10 }
{ userName: 'Pooja', value: 10 }
finished parse

NOTE: A list of types that can be passed can be found on the Exceljs project

Writing Workbooks

Workbooks must be given a schema for each sheet before it can be written to. Data is then added to the workbook in the form of json. All this data is saved to the sheet using the save function, which returns a promise which resolves the stream. This stream can then be piped or written to the fileSystem using the fs module.

Input: var inputs

[
    {name: 'Test 1', testValue: 100},
    {name: 'Test 2'},
    {name: 'Test 3', testValue: 80}
]

Writing:

let writer = new ExcelWriter({
    sheets: [{
        name: 'Test Sheet',
        key: 'tests',
        headers: [{
            name: 'Test Name',
            key: 'name'
        }, {
            name: 'Test Coverage',
            key: 'testValue',
            default: 0
        }]
    }]
});
let dataPromises = inputs.map((input) => {
    // 'tests' is the key of the sheet. That is used
    // to add data to only the Test Sheet
    writer.addData('tests', input);
});
Promise.all(dataPromises)
.then(() => {
    return writer.save();
})
.then((stream) => {
    stream.pipe(fs.createWriteStream('data.xlsx'));
});

Output: Input: data.xlsx, Sheet: Test Sheet

Test Name	Test Coverage
Test 1	100
Test 2	0
Test 3	80

Configuration

Reading

node-excel-stream takes a schema to strictly check the excel format while reading. This makes sure that the excel contains data in the correct format, and throws errors when there is a mismatch.

Configuration options -

{
    sheets: [{
        name: 'Users',              // The name of the sheet. Required
        key: 'users',               // The key of the sheet. Optional. Used to identify the sheet while getting the row data
        rows: {
            headerRow: 1,           // The row that contains the headers. Rows above this are ignored. Default: 1
            allowedHeaders: [{
                name: 'User Name',  // The text value of the header cell
                key: 'userName',    // The identifier for the header. Will be the field name in the row data json
                type: String        // The type, for strict type checking for the data
            }]
        }
    }]
}

Contributing

PRs are always welcome. Thank you to all contributors :)