@csv-streamy/lib
TypeScript icon, indicating that this package has built-in type declarations

1.0.7 • Public • Published

@csv-streamy/lib

npm version CI codecov License: MIT

CSV Streamy Lib - CSV Stream library for Node.js.

Install

npm i @csv-streamy/lib

Quick examples

Read and Write your CSV file

import { resolve } from 'path'
import { pipeline } from 'stream/promises'
import { createReadStream, createWriteStream } from 'fs'
import { createCsvParser, createCsvConverter, CsvRowData } from '@csv-streamy/lib'

// For example, it just converts all fields to uppercase letters.
async function processRow({ data }: CsvRowData): Promise<CsvRowData> {
  for (const [header, field] of Object.entries(data)) {
    data[header] = field.toUpperCase()
  }
  return Promise.resolve({ data })
}

async function run() {
  await pipeline(
    createReadStream(resolve(__dirname, 'input.csv'), { encoding: 'utf-8' }),
    createCsvParser({ hasHeaders: true, hasDoubleQuotes: true }),
    async function* (source) {
      for await (const row of source) {
        yield await processRow(row as CsvRowData)
      }
    },
    createCsvConverter({ hasHeaders: true, hasDoubleQuotes: true }),
    createWriteStream(resolve(__dirname, 'output.csv'), { encoding: 'utf-8' }),
  )
  console.log('Woo-hoo! Succeeded!!')
}

run().catch(console.error)

Read and Write your CSV file with stat

import { resolve } from 'path'
import { pipeline } from 'stream/promises'
import { createReadStream, createWriteStream } from 'fs'
import { createCsvParser, createCsvConverter, CsvRowData } from '@csv-streamy/lib'

// You can observe the number of row as `count` and the bytes of data as `amount` in `stat`.
// For example, it converts all fields to uppercase letters if `count` is even
// or it capitalizes all fields if `amount` is 200 bytes or more.
async function processRow({ data, stat }: CsvRowData): Promise<CsvRowData> {
  const { count, amount } = { ...stat }
  if (!!count && count % 2 === 0) {
    for (const [header, field] of Object.entries(data)) {
      data[header] = field.toUpperCase()
    }
  } else if (!!amount && amount >= 200) {
    for (const [header, field] of Object.entries(data)) {
      data[header] = field.charAt(0).toUpperCase() + field.slice(1)
    }
  }
  return Promise.resolve({ data })
}

async function run() {
  await pipeline(
    createReadStream(resolve(__dirname, 'input.csv'), { encoding: 'utf-8' }),
    createCsvParser({ hasHeaders: true, hasDoubleQuotes: true }),
    async function* (source) {
      for await (const row of source) {
        yield await processRow(row as CsvRowData)
      }
    },
    createCsvConverter({ hasHeaders: true, hasDoubleQuotes: true }),
    createWriteStream(resolve(__dirname, 'output.csv'), { encoding: 'utf-8' }),
  )
  console.log('Great! Perfect!!')
}

run().catch(console.error)

Using ES Modules (ESM)

If you want to use ES Modules, you can do it as follows.

import { resolve } from 'path'
import { dirname } from 'dirfilename'
import { pipeline } from 'stream/promises'
import { createReadStream, createWriteStream } from 'fs'
import { createCsvParser, createCsvConverter, CsvRowData } from '@csv-streamy/lib'

// Workaround to simply use `__dirname` because CommonJS variables are not available in ES modules.
const __dirname = dirname(import.meta.url)

async function processRow({ data }: CsvRowData): Promise<CsvRowData> {
  for (const [header, field] of Object.entries(data)) {
    data[header] = field.toUpperCase()
  }
  return Promise.resolve({ data })
}

async function run() {
  await pipeline(
    createReadStream(resolve(__dirname, 'input.csv'), { encoding: 'utf-8' }),
    createCsvParser({ hasHeaders: true, hasDoubleQuotes: true }),
    async function* (source) {
      for await (const row of source) {
        yield await processRow(row as CsvRowData)
      }
    },
    createCsvConverter({ hasHeaders: true, hasDoubleQuotes: true }),
    createWriteStream(resolve(__dirname, 'output.csv'), { encoding: 'utf-8' }),
  )
  console.log('Woo-hoo! Succeeded!!')
}

run().catch(console.error)

Using CommonJS

If you want to use CommonJS in just Node.js, you can do it as follows.

const { resolve } = require('path')
const { pipeline } = require('stream/promises')
const { createReadStream, createWriteStream } = require('fs')
const { createCsvParser, createCsvConverter } = require('@csv-streamy/lib')

async function processRow({ data }) {
  for (const [header, field] of Object.entries(data)) {
    data[header] = field.toUpperCase()
  }
  return Promise.resolve({ data })
}

async function run() {
  await pipeline(
    createReadStream(resolve(__dirname, 'input.csv'), { encoding: 'utf-8' }),
    createCsvParser({ hasHeaders: true, hasDoubleQuotes: true }),
    async function* (source) {
      for await (const row of source) {
        yield await processRow(row)
      }
    },
    createCsvConverter({ hasHeaders: true, hasDoubleQuotes: true }),
    createWriteStream(resolve(__dirname, 'output.csv'), { encoding: 'utf-8' })
  );
  console.log('Woo-hoo! Succeeded!!')
}

run().catch(console.error)

Usage

Parsing

You can parse your csv file, which can contain headers and enclose fields in double-quotes, to handy objects. Each object contains fields per row as data and statistics data as stat, which contains the number of row as count and the bytes of data as amount.

import { resolve } from 'path'
import { createReadStream } from 'fs'
import { createCsvParser } from '@csv-streamy/lib'

const reader = createReadStream(resolve(__dirname, 'input.csv'))
const parser = createCsvParser({ hasHeaders: true, hasDoubleQuotes: true })

reader
  .pipe(parser)
  .on('error', (error) => console.log(error))
  .on('data', (row) => console.log(row))
  .on('end', () => console.log('End'))
  • Input
"header[1]","header[2]","header[3]","header[4]","header[5]"
"item[1][1]","item[1][2]","item[1][3]","item[1][4]","item[1][5]"
"item[2][1]","item[2][2]","item[2][3]","item[2][4]","item[2][5]"
  • Output
{
  data: {
    'header[1]': 'item[1][1]',
    'header[2]': 'item[1][2]',
    'header[3]': 'item[1][3]',
    'header[4]': 'item[1][4]',
    'header[5]': 'item[1][5]'
  },
  stat: { count: 1, amount: 64 }
}
{
  data: {
    'header[1]': 'item[2][1]',
    'header[2]': 'item[2][2]',
    'header[3]': 'item[2][3]',
    'header[4]': 'item[2][4]',
    'header[5]': 'item[2][5]'
  },
  stat: { count: 2, amount: 128 }
}
End

Converting

You can convert your handy objects with data to csv format strings, which can contain headers and enclose fields in double-quotes, then create buffer stream.

import { createCsvConverter } from '@csv-streamy/lib'

const converter = createCsvConverter({ hasHeaders: true, hasDoubleQuotes: true })

converter.pipe(process.stdout).on('end', () => process.exit())

converter.write({ data: { 'header[1]': 'item[1][1]', 'header[2]': 'item[1][2]', 'header[3]': 'item[1][3]' } })
converter.write({ data: { 'header[1]': 'item[2][1]', 'header[2]': 'item[2][2]', 'header[3]': 'item[2][3]' } })
converter.write({ data: { 'header[1]': 'item[3][1]', 'header[2]': 'item[3][2]', 'header[3]': 'item[3][3]' } })
converter.end()
  • Output
"header[1]","header[2]","header[3]"
"item[1][1]","item[1][2]","item[1][3]"
"item[2][1]","item[2][2]","item[2][3]"
"item[3][1]","item[3][2]","item[3][3]"

Of course, you can export them to a file.

import { resolve } from 'path'
import { createWriteStream } from 'fs'
import { createCsvConverter } from '@csv-streamy/lib'

const converter = createCsvConverter({ hasHeaders: true, hasDoubleQuotes: true })
const writer = createWriteStream(resolve(__dirname, 'output.csv'))

converter.pipe(writer).on('end', () => writer.end())

converter.write({ data: { 'header[1]': 'item[1][1]', 'header[2]': 'item[1][2]', 'header[3]': 'item[1][3]' } })
converter.write({ data: { 'header[1]': 'item[2][1]', 'header[2]': 'item[2][2]', 'header[3]': 'item[2][3]' } })
converter.write({ data: { 'header[1]': 'item[3][1]', 'header[2]': 'item[3][2]', 'header[3]': 'item[3][3]' } })
converter.end()

Acceptable CSV format

This basically follows RFC4180 but additionally needs to meet the following rules to make a common csv file easier to use:

  • Fields must be Comma-Separated Values. (NOT Tab-Separated.)
  • Fields can be enclosed in double-quotes to contain line breaks, double quotes and commas. (BUT a file cannot mix enclosed fields and not-enclosed fields.)
  • A double-quote appearing inside a field must be escaped by preceding it with another double-quote or a backslash.
  • This doesn't check a MIME Type such as text/csv.

Package Sidebar

Install

npm i @csv-streamy/lib

Weekly Downloads

0

Version

1.0.7

License

MIT

Unpacked Size

38.5 kB

Total Files

23

Last publish

Collaborators

  • keidrun