csv-load-sync v2.3.1

csv-load-sync

Sync loading routine for a small comma-separated values file (.csv). Returns an array of objects, takes property names from the first line. Assumes everything is a string in quotes.

Install

Requires nodejs

npm install -S csv-load-sync

Inputs

Example CSV file with two records

"deviceId","description"
"1","iPhone 4"
"2","iPhone 4S"

Example CSV file with three records

"id","firstName","lastName","country","lastLogin"
"1","John","Smith","US","2013-08-04 23:57:38"
"2","Greg","Smith","US","2013-07-12 13:27:18"
"3","Harold","Smith","GB","2013-07-16 21:17:28"

Simple example

const {load} = require('csv-load-sync');
const csv = load('path/to/file.csv');
// csv is an Array of objects

Convert values

You can convert every value from string to a desired type for the specified columns. For example to convert device IDs to an integer, while make the descriptions all uppercase:

const {load} = require('csv-load-sync')
const csv = load('./phone.csv', {
  convert: {
    deviceId: parseInt,
    description: (s) => s.toUpperCase(),
  },
})
// objects:
// deviceId: 1, description: "IPHONE 4"
// deviceId: 2, description: "IPHONE 4S"

Useful conversions:

• to boolean: (s) => (s === 'true' ? true : false)
• to integer: parseInt
• to floats: parseFloat

Custom line splitting

Sometimes CSV data includes commas naturally, for example the follwing file has GPS location numbers which should be considered together.

"place","location"
"home",-41.20,20.11
"work",-41.3,20.2

We need to split each record line differently. You can pass the line to columns splitter function when calling load

function split(line, lineNumber) {
  if (lineNumber === 0) { // title line
    return line.split(',')
  }
  // our line will be <location>,<lat>,<lon>
  // and we want to combine lat and lon
  var parts = line.split(',')
  return [parts[0], parts[1] + ',' + parts[2]];
}
var results = load(filename, {
  getColumns: split
});
/*
[{
  place: 'home',
  location: '-41.20,20.11'
}, {
  place: 'work',
  location: '-41.3,20.2'
}]
*/

Comments

Blank lines and lines starting with # are skipped.

Skip columns

Sometimes you want to skip certain columns. Use an option:

const results = load(filename, {
  skip: ['lastLogin', 'country'],
})

Parse given text

If you load the CSV text yourself, you can use to parse it

const {parseCSV} = require('csv-load-sync')
// csv is your text to be parsed
const records = parseCSV(csv)

Get columns

You can just get the header names from the CSV file

const {getHeaders} = require('csv-load-sync')
const names = getHeaders('path/to/file.csv')
// names is an array of strings

Fine print

Author: Gleb Bahmutov © 2015

• @bahmutov
• glebbahmutov.com
• blog
• video channel

License: MIT - do anything with the code, but don't blame me if it does not work.

Spread the word: tweet, star on github, etc.