CSV.js

Simple, ultra light (10kb uncompressed) javascript CSV library for browser and node with zero dependencies.

Originally developed as part of ReclineJS but now fully standalone.

Usage

Grab the csv.js file and include it in your application.

Depends on jQuery or underscore.deferred (for deferred) in fetch (and jQuery if you need ajax). parse and serialize have zero dependencies.

fetch

A convenient way to load a CSV file from various different sources. fetch supports 3 options depending on the attribute provided on the info argument:

CSV.fetch({
    data: 'raw csv string'
    // or ...
    url: 'url to a csv file'
    // or ...
    file: an HTML 5 file object

    // optional options about structure of the CSV file
    // following the CSV Dialect Description Format 
    // https://frictionlessdata.io/specs/csv-dialect/
    dialect: {
      ...
    }
  }
).done(function(dataset) {
  // dataset object doc'd below
  console.log(dataset);
});

Some more detail on the argument object:

• data is a string in CSV format. This is passed directly to the CSV parser
• url: a url to an online CSV file that is ajax accessible (note this usually requires either local or on a server that is CORS enabled). The file is then loaded using jQuery.ajax and parsed using the CSV parser (NB: this requires jQuery) All options generates similar data and use the memory store outcome, that is they return something like:
• file: is an HTML5 file object. This is opened and parsed with the CSV parser.
• dialect: hash / dictionary following the same structure as for parse method below.

Returned dataset object looks like:

Raw parsing

var out = CSV.parse(csvString, dialect);

Converts a Comma Separated Values string into an array of arrays. Each line in the CSV becomes an array.

Empty fields are converted to nulls and non-quoted numbers are converted to integers or floats.

• csvString: the csv string to parse

• dialect: [optional] hash with keys as per the CSV dialect description format. It also supports the following additional keys:

  • skipInitialRows: [optional] integer number of rows to skip (default 0)

  For backwards compatability with earlier versions of the library the dialect also supports the following:

  • trim: mapped to skipInitialSpace in CSV dialect description format

Serialize

Convert an Object or a simple array of arrays into a Comma Separated Values string.

var out = CSV.serialize(dataToSerialize, dialect);

Returns a string representing the array serialized as a CSV.

dataToSerialize is an Object or array of arrays to convert. Object structure must be as follows:

{
  fields: [ {id: .., ...}, {id: ..., 
  records: [ { record }, { record }, ... ]
  ... // more attributes we do not care about
}

Nulls are converted to empty fields and integers or floats are converted to non-quoted numbers.

You may optionally specify a label inside each field so that the serialized data will use it as the column heading instead of the id.

• dialect: dialect options for serializing the CSV file as per CSV Dialect Description Format

Other JS CSV Libs

• http://www.uselesscode.org/javascript/csv/ - basic CSV parser on which this library was originally based
• https://github.com/maxogden/browser-csv-stream - Pure browser version of node-csv from @maxogden via browserify
• https://github.com/onyxfish/csvkit.js - pure JS CSV reader from @onyxfish (author of the "legendary" python csvkit)
• https://github.com/mholt/PapaParse - fast CSV parser that can handle large files and malformed data

Node

• https://github.com/wdavidw/node-csv - this is the Node CSV lib we use by preference
• https://github.com/maxogden/binary-csv - new CSV lib from @maxogden with a focus on being very fast

Development

Requirements

• webpack
• jquery

npm install
npm install jquery
webpack-dev-server

Run tests

Requirements

• karma
• phantomjs

npm -g install karma karma-cli phantomjs-prebuilt
npm install
npm install jquery
npm test