A JavaScript library for internationalization and localization that leverage the official Unicode CLDR JSON data. The library works both for the browser and as a Node.js module.
Each language, and the countries that speak that language, have different expectations when it comes to how numbers (including currency and percentages) and dates should appear. Obviously, each language has different names for the days of the week and the months of the year. But they also have different expectations for the structure of dates, such as what order the day, month and year are in. In number formatting, not only does the character used to delineate number groupings and the decimal portion differ, but the placement of those characters differ as well.
A user using an application should be able to read and write dates and numbers in the format they are accustomed to. This library makes this possible, providing an API to convert user-entered number and date strings - in their own format - into actual numbers and dates, and conversely, to format numbers and dates into that string format.
Even if the application deals only with the English locale, it may still need globalization to format programming language bytes into human-understandable language and vice-versa in an effective and reasonable way. For example, to display something better than "Edited 1 minutes ago".
Globalize provides number formatting and parsing, date and time formatting and parsing, currency formatting, message formatting (ICU message format pattern), and plural support.
Design Goals.
- Leverages the Unicode CLDR data and follows its UTS#35 specification.
- Keeps code separate from i18n content. Doesn't host or embed any locale data in the library. Empowers developers to control the loading mechanism of their choice.
- Allows developers to load as much or as little data as they need. Avoids duplicating data if using multiple i18n libraries that leverage CLDR.
- Keeps code modular. Allows developers to load the i18n functionalities they need.
- Runs in browsers and Node.js, consistently across all of them.
- Makes globalization as easy to use as jQuery.
Globalize is based on the Unicode Consortium's Common Locale Data Repository (CLDR), the largest and most extensive standard repository of locale data available. CLDR is constantly updated and is used by many large applications and operating systems, so you'll always have access to the most accurate and up-to-date locale data.
Globalize needs CLDR content to function properly, although it doesn't embed, hard-code, or host such content. Instead, Globalize empowers developers to load CLDR data the way they want. Vanilla CLDR in its official JSON format (no pre-processing) is expected to be provided. As a consequence, (a) Globalize avoids bugs caused by outdated i18n content. Developers can use up-to-date CLDR data directly from Unicode as soon as it's released, without having to wait for any pipeline on our side. (b) Developers have full control over which locale coverage they want to provide on their applications. (c) Developers are able to share the same i18n dataset between Globalize and other libraries that leverage CLDR. There's no need for duplicating data.
Globalize is systematically tested against desktop and mobile browsers and Node.js. So, using it you'll get consistent results across different browsers and across browsers and the server.
Globalize doesn't use native Ecma-402 yet, which could potentially improve date and number formatting performance. Although Ecma-402 support is improving among modern browsers and even Node.js, the functionality and locale coverage level varies between different environments (see Comparing JavaScript Libraries slide 25). Globalize needs to do more research and testings to use it reliably.
For alternative libraries and more, check out this JavaScript globalization overview.
Are you coming from Globalize 0.x? Read our migration guide to learn what have changed and how to migrate older 0.x code to up-to-date 1.x.
Globalize is designed to work both in the browser, or in Node.js. It supports both AMD and CommonJS.
Globalize uses the Unicode CLDR, the largest and most extensive standard repository of locale data.
We do NOT embed any i18n data within our library. However, we make it really easy to use. Read How to get and load CLDR JSON data for more information on its usage.
File | Minified gzipped size | Runtime minified gzipped size | Summary |
---|---|---|---|
globalize.js | 1.7KB | 1.1KB | Core library |
globalize/currency.js | 3.0KB | 0.7KB | Currency module provides currency formatting |
globalize/date.js | 7.7KB | 4.3KB | Date module provides date formatting and parsing |
globalize/message.js | 5.3KB | 0.7KB | Message module provides ICU message format support |
globalize/number.js | 4.4KB | 2.6KB | Number module provides number formatting and parsing |
globalize/plural.js | 2.3KB | 0.4KB | Plural module provides pluralization support |
globalize/relative-time.js | 0.8KB | 0.5KB | Relative time module provides relative time formatting support |
globalize/unit.js | 0.9KB | 0.6KB | Unit module provides unit formatting support |
Globalize 1.x supports the following browsers:
- Chrome: (Current - 1) or Current
- Firefox: (Current - 1) or Current
- Safari: 5.1
- Opera: 12.1x, (Current - 1) or Current
- IE9
(Current - 1) or Current denotes that we support the current stable version of the browser and the version that preceded it. For example, if the current version of a browser is 24.x, we support the 24.x and 23.x versions.
npm install globalize cldr-data iana-tz-data
var Globalize = require( "globalize" );
Globalize.load( require( "cldr-data" ).entireSupplemental() );
Globalize.load( require( "cldr-data" ).entireMainFor( "en", "es" ) );
Globalize.loadTimeZone( require( "iana-tz-data" ) );
Globalize("en").formatDate(new Date());
// > "11/27/2015"
Globalize("es").formatDate(new Date());
// > "27/11/2015"
Note cldr-data
is an optional module, read CLDR content section below for more information on how to get CLDR from different sources.
The iana-tz-data
module is only needed when IANA time zones (via options.timeZone
) are used with date functions. Read IANA time zone data below for more information.
Read the Locales section for more information about supported locales. For AMD, bower and other usage examples, see Examples section.
Click the GitHub releases tab and download the latest available Globalize package.
You can use either npm or bower:
npm install globalize
bower install globalize
git clone https://github.com/globalizejs/globalize.git
- Build the distribution files
If you use module loading like ES6 import, CommonJS, or AMD and fetch your code using package managers like npm or bower, you don't need to worry about this and can skip reading this section. Otherwise, you need to satisfy Globalize dependencies prior to using it. There is only one external dependency: cldr.js, which is a CLDR low level manipulation tool. Additionally, you need to satisfy the cross-dependencies between modules.
Module | Dependencies (load in order) |
---|---|
Core module | cldr.js |
Currency module | globalize.js (core), globalize/number.js, and globalize/plural.js (only required for "code" or "name" styles) |
Date module | globalize.js (core) and globalize/number.js |
Message module | globalize.js (core) and globalize/plural.js (if using messages that need pluralization support) |
Number module | globalize.js (core) |
Plural | globalize.js (core) |
Relative time module | globalize.js (core), globalize/number.js, and globalize/plural.js |
Unit module | globalize.js (core), globalize/number.js, and globalize/plural.js |
As an alternative to deducing this yourself, use this online tool. The tool allows you to select the modules you're interested in using and tells you the Globalize files and CLDR JSON that you need.
Globalize is the i18n software (the engine). Unicode CLDR is the i18n content (the fuel). You need to feed Globalize on the appropriate portions of CLDR prior to using it.
(a) How do I figure out which CLDR portions are appropriate for my needs?
Each Globalize function requires a special set of CLDR portions. Once you know which Globalize functionalities you need, you can deduce its respective CLDR requirements. See table below.
Module | Required CLDR JSON files |
---|---|
Core module | cldr/supplemental/likelySubtags.json |
Currency module | cldr/main/locale /currencies.jsoncldr/supplemental/currencyData.json CLDR JSON files from number module CLDR JSON files from plural module for name style support |
Date module | cldr/main/locale /ca-gregorian.jsoncldr/main/ locale /timeZoneNames.jsoncldr/supplemental/metaZones.json cldr/supplemental/timeData.json cldr/supplemental/weekData.json CLDR JSON files from number module |
Number module | cldr/main/locale /numbers.jsoncldr/supplemental/numberingSystems.json |
Plural module | cldr/supplemental/plurals.json (for cardinals) cldr/supplemental/ordinals.json (for ordinals) |
Relative time module | cldr/main/locale /dateFields.jsonCLDR JSON files from number and plural modules |
Unit module | cldr/main/locale /units.jsonCLDR JSON files from number and plural module |
As an alternative to deducing this yourself, use this online tool. The tool allows you to select the modules you're interested in using and tells you the Globalize files and CLDR JSON that you need.
(b) How am I supposed to get and load CLDR content?
Learn how to get and load CLDR content... and use
Globalize.load()
to load it.
The IANA time zone (tz) database, sometimes called the Olson database, is the standard data used by Unicode CLDR, ECMA-402, Linux, UNIX, Java, ICU, and others. It's used by Globalize to circumvent the JavaScript limitations with respect to manipulating date in time zones other than the user's environment.
In short, feed Globalize on IANA time zone data if you need to format or parse dates in a specific time zone, independently of the user's environment, e.g., America/Los_Angeles
.
It's important to note there's no official IANA time zone data in the JSON format. Therefore, iana-tz-data
has been adopted for convenience.
Learn more on Globalize.loadTimeZone()
.
Globalize's consumable-files are located in the ./dist
directory. If you don't find it, it's because you are using a development branch. You should either use a tagged version or build the distribution files yourself. Read installation above if you need more information on how to download.
Globalize can be used for a variety of different i18n tasks, eg. formatting or parsing dates, formatting or parsing numbers, formatting messages, etc. You may NOT need Globalize in its entirety. For that reason, we made it modular. So, you can cherry-pick the pieces you need, eg. load dist/globalize.js
to get Globalize core, load dist/globalize/date.js
to extend Globalize with Date functionalities, etc.
An example is worth a thousand words. Check out our Examples section below.
When formatting or parsing, there's actually a two-step process: (a) the formatter (or parser) creation and (b) its execution, where creation takes an order of magnitude more time (more expensive) than execution. In the creation phase, Globalize traverses the CLDR tree, processes data (e.g., expands date patterns, parses plural rules, etc), and returns a function that actually executes the formatting or parsing.
// Formatter creation.
var formatter = Globalize.numberFormatter();
// Formatter execution (roughly 10x faster than above).
formatter( Math.PI );
// > 3.141
As a rule of thumb for optimal performance, cache your formatters and parsers. For example: (a) on iterations, generate them outside the loop and reuse while looping; (b) on server applications, generate them in advance and execute when requests arrive.
Take advantage of compiling your formatters and/or parsers during build time when deploying to production. It's much faster than generating them in real-time and it's also much smaller (i.e., better loading performance).
Your compiled formatters and parsers allow you to skip a big part of the library and also allow you to skip loading CLDR data, because they have already been created (see Performance above for more information).
To illustrate, see our Basic Globalize Compiler example.
For information about the Globalize Compiler CLI or its JavaScript API, see the Globalize Compiler documentation.
The fastest and easiest way to use Globalize is by integrating it into your existing tools.
- Application example using webpack and npm: easy to get started, automated CLDR loading and precompilation for production, but requires npm and webpack knowledge.
- Application example using globalize-express middleware with any express web app: easy to incorporate globalize as a middleware within your Express web app. (also checkout globalize-express)
If you're using a different tool than the one above, but you're comfortable using JavaScript modules (such as ES6 modules, CommonJS, or AMD) and package managers like npm or bower, you may want to check out the following examples. Note you'll need to compile your code for production yourself.
- Basic example using AMD and bower: feeding Globalize on CLDR is not completely transparent.
- Basic example using Node.js and npm: feeding Globalize on CLDR is not completely transparent.
- Basic Globalize Compiler example: shows how to use Globalize Compiler CLI.
If you're using none of the tools above, but instead you're using the plain and old script tags only, the following example may interest you. Note Globalize allows you to go low level like this. But, acknowledge that you'll need to handle dependencies and CLDR loading manually yourself.
- Basic example using plain JavaScript: requires loading CLDR and handling dependencies manually.
You can find us on Slack. If you're new, join here.
This method allows you to load CLDR JSON locale data. Globalize.load()
is a proxy to Cldr.load()
. Read more...
Set default locale, or get it if locale argument is omitted. Read more...
Create a Globalize instance. Read more...
A locale is an identifier (id) that refers to a set of user preferences that tend to be shared across significant swaths of the world. In technical terms, it's a String composed of three parts: language, script, and region. For example:
locale | description |
---|---|
en-Latn-US | English as spoken in the Unites States in the Latin script. |
en-US | English as spoken in the Unites States (Latin script is deduced given it's the most likely script used in this place). |
en | English (United States region and Latin script are deduced given they are respectively the most likely region and script used in this place). |
en-GB | English as spoken in the United Kingdom (Latin script is deduced given it's the most likely script used in this place). |
en-IN | English as spoken in India (Latin script is deduced). |
es | Spanish (Spain region and Latin script are deduced). |
es-MX | Spanish as spoken in Mexico (Latin script is deduced). |
zh | Chinese (China region and Simplified Han script are deduced). |
zh-TW | Chinese as spoken in Taiwan (Traditional Han script is deduced). |
ja | Japanese (Japan region and Japanese script are deduced). |
de | German (Germany region and Latin script are deduced). |
pt | Portuguese (Brazil region and Latin script are deduced). |
pt-PT | Portuguese as spoken in Portugal (Latin script is deduced). |
fr | French (France region and Latin script are deduced). |
ru | Russian (Russia region and Cyrillic script are deduced). |
ar | Arabic (Egypt region and Arabic script are deduced). |
The likely deductibility is computed by using CLDR data, which is based on the population and the suppress-script data in BCP47 (among others). The data is heuristically derived, and may change over time.
Figure out the deduced information by looking at the cldr.attributes.maxLanguageId
property of a Globalize instance:
var Globalize = require( "globalize" );
Globalize.load( require( "cldr-data" ).entireSupplemental() );
Globalize( "en" ).cldr.attributes.maxLanguageId;
// > "en-Latn-US"
Globalize supports all the locales available in CLDR, which are around 740. For more information, search for coverage charts at the downloads section of http://cldr.unicode.org/.
Read more details about locale at UTS#35 locale.
This method allows you to load IANA time zone data to enable options.timeZone
feature on date formatters and parsers.
Return a function that formats a date according to the given options
. The default formatting is numeric year, month, and day (i.e., { skeleton: "yMd" }
.
.dateFormatter()( new Date() )
// > "11/30/2010"
.dateFormatter({ skeleton: "GyMMMd" })( new Date() )
// > "Nov 30, 2010 AD"
.dateFormatter({ date: "medium" })( new Date() )
// > "Nov 1, 2010"
.dateFormatter({ time: "medium" })( new Date() )
// > "5:55:00 PM"
.dateFormatter({ datetime: "medium" })( new Date() )
// > "Nov 1, 2010, 5:55:00 PM"
.dateFormatter({ datetime: "full", timeZone: "America/New_York" })( new Date() );
// > "Monday, November 1, 2010 at 3:55:00 AM Eastern Daylight Time"
.dateFormatter({ datetime: "full", timeZone: "America/Los_Angeles" })( new Date() );
// > "Monday, November 1, 2010 at 12:55:00 AM Pacific Daylight Time"
Return a function that formats a date into parts tokens according to the given options
. The default formatting is numeric year, month, and day (i.e., { skeleton: "yMd" }
.
.dateToPartsFormatter()( new Date() )
// > [
// { "type": "month", "value": "3" },
// { "type": "literal", "value": "/" },
// { "type": "day", "value": "17" },
// { "type": "literal", "value": "/" },
// { "type": "year", "value": "2017" }
// ]
Return a function that parses a string representing a date into a JavaScript Date object according to the given options
. The default parsing assumes numeric year, month, and day (i.e., { skeleton: "yMd" }
).
.dateParser()( "11/30/2010" )
// > new Date( 2010, 10, 30, 0, 0, 0 )
.dateParser({ skeleton: "GyMMMd" })( "Nov 30, 2010 AD" )
// > new Date( 2010, 10, 30, 0, 0, 0 )
.dateParser({ date: "medium" })( "Nov 1, 2010" )
// > new Date( 2010, 10, 30, 0, 0, 0 )
.dateParser({ time: "medium" })( "5:55:00 PM" )
// > new Date( 2015, 3, 22, 17, 55, 0 ) // i.e., today @ 5:55PM
.dateParser({ datetime: "medium" })( "Nov 1, 2010, 5:55:00 PM" )
// > new Date( 2010, 10, 30, 17, 55, 0 )
Alias for .dateFormatter( [options] )( value )
.
Alias for .dateToPartsFormatter( [options] )( value )
.
Alias for .dateParser( [options] )( value )
.
Load messages data. Read more...
Return a function that formats a message (using ICU message format pattern) given its path and a set of variables into a user-readable string. It supports pluralization and gender inflections.
.messageFormatter( "task" )( 1000 )
// > "You have 1,000 tasks remaining"
.messageFormatter( "like" )( 3 )
// > "You and 2 others liked this"
Alias for .messageFormatter( path )([ variables ])
.
Return a function that formats a number according to the given options or locale's defaults.
.numberFormatter()( pi )
// > "3.142"
.numberFormatter({ maximumFractionDigits: 5 })( pi )
// > "3.14159"
.numberFormatter({ round: "floor" })( pi )
// > "3.141"
.numberFormatter({ minimumFractionDigits: 2 })( 10000 )
// > "10,000.00"
.numberFormatter({ style: "percent" })( 0.5 )
// > "50%"
.numberFormatter({ compact: "short", maximumFractionDigits: 0 })( 14305 )
// > "14K"
Return a function that formats a number into parts tokens according to the given options or locale's defaults.
.numberToPartsFormatter()( new Date() )
// > [
// { "type": "integer", "value": "3" },
// { "type": "decimal", "value": "." },
// { "type": "fraction", "value": "142" }
// ]
Return a function that parses a string representing a number according to the given options or locale's defaults.
.numberParser()( "3.14159" )
// > 3.14159
.numberParser()( "10,000.00" )
// > 10000
.numberParser({ style: "percent" })( "50%" )
// > 0.5
Alias for .numberFormatter( [options] )( value )
.
Alias for .numberToPartsFormatter( [options] )( value )
.
Alias for .numberParser( [options] )( value )
.
Return a function that formats a currency according to the given options or locale's defaults.
.currencyFormatter( "USD" )( 1 )
// > "$1.00"
.currencyFormatter( "USD", { style: "accounting" })( -1 )
// > "($1.00)"
.currencyFormatter( "USD", { style: "name" })( 69900 )
// > "69,900.00 US dollars"
.currencyFormatter( "USD", { style: "code" })( 69900 )
// > "69,900.00 USD"
.currencyFormatter( "USD", { round: "ceil" })( 1.491 )
// > "$1.50"
Return a function that formats a currency into parts tokens according to the given options or locale's defaults.
.currencyToPartsFormatter()( new Date() )
// > [
// { "type": "currency", "value": "USD" },
// { "type": "literal", "value": " " },
// { "type": "integer", "value": "69" },
// { "type": "group", "value": "," },
// { "type": "integer", "value": "900" },
// { "type": "decimal", "value": "." },
// { "type": "fraction", "value": "00" }
// ]
Alias for .currencyFormatter( currency [, options] )( value )
.
Alias for .currencyToPartsFormatter( currency [, options] )( value )
.
Return a function that returns the value's corresponding plural group: zero
, one
, two
, few
, many
, or other
.
The function may be used for cardinals or ordinals.
.pluralGenerator()( 0 )
// > "other"
.pluralGenerator()( 1 )
// > "one"
.pluralGenerator({ type: "ordinal" })( 1 )
// > "one"
.pluralGenerator({ type: "ordinal" })( 2 )
// > "two"
Alias for .pluralGenerator( [options] )( value )
.
Returns a function that formats a relative time according to the given unit, options, and the default/instance locale.
.relativeTimeFormatter( "day" )( 1 )
// > "tomorrow"
.relativeTimeFormatter( "month" )( -1 )
// > "last month"
.relativeTimeFormatter( "month" )( 3 )
// > "in 3 months"
Alias for .relativeTimeFormatter( unit, options )( value )
.
Returns a function that formats a unit according to the given unit, options, and the default/instance locale.
.unitFormatter( "second" )( 10 )
// > "10 seconds"
.unitFormatter( "second", { form: "short" } )( 10 )
// > "10 secs"
.unitFormatter( "second", { form: "narrow" } )( 10 )
// > "10s"
Alias for .unitFormatter( unit, options )( value )
.
-
E_INVALID_CLDR
Thrown when a CLDR item has an invalid or unexpected value.
-
E_MISSING_CLDR
Thrown when any required CLDR item is NOT found.
-
E_INVALID_PAR_TYPE
Thrown when a parameter has an invalid type on any static or instance methods.
-
E_INVALID_PAR_VALUE
Thrown for certain parameters when the type is correct, but the value is invalid.
-
E_MISSING_PARAMETER
Thrown when a required parameter is missing on any static or instance methods.
-
E_PAR_OUT_OF_RANGE
Thrown when a parameter is not within a valid range of values.
-
E_DEFAULT_LOCALE_NOT_DEFINED
Thrown when any static method, eg.
Globalize.formatNumber()
is used prior to setting the Global locale withGlobalize.locale( <locale> )
. -
E_MISSING_PLURAL_MODULE
Thrown when plural module is needed, but not loaded, eg. to format currencies using the named form.
-
E_UNSUPPORTED
Thrown for unsupported features, eg. to format unsupported date patterns.
If you are having trouble using Globalize after reading the documentation carefully, please post a question to StackOverflow with the "javascript-globalize" tag. Questions that include a minimal demo are more likely to receive responses.
In the spirit of open source software development, we always encourage community code contribution. To help you get started and before you jump into writing code, be sure to read CONTRIBUTING.md.
For ideas where to start contributing, see the following queries to find what best suites your interest: quick change, new features, bug fixes, documentation improvements, date module, currency module, message module, number module, plural module, relative time module. Last but not least, feel free to get in touch.
Our roadmap is the collection of all open issues and pull requests where you can find:
- Ongoing work lists our current sprint. Here you find where we're actively working on at this very moment. Priority is determined by the community needs and volunteering. If there is anything you want to be done, share your thoughts with us on any existing or new issue and especially volunteer to do it.
- Everything else is potential next work that you could help us to accomplish now. Releases are published following semver rules as often as possible.
├── bower.json (metadata file)
├── CONTRIBUTING.md (doc file)
├── dist/ (consumable files, the built files)
├── external/ (external dependencies, eg. cldr.js, QUnit, RequireJS)
├── Gruntfile.js (Grunt tasks)
├── LICENSE (license file)
├── package.json (metadata file)
├── README.md (doc file)
├── src/ (source code)
│ ├── build/ (build helpers, eg. intro, and outro)
│ ├── common/ (common function helpers across modules)
│ ├── core.js (core module)
│ ├── date/ (date source code)
│ ├── date.js (date module)
│ ├── message.js (message module)
│ ├── number.js (number module)
│ ├── number/ (number source code)
│ ├── plural.js (plural module)
│ ├── plural/ (plural source code)
│ ├── relative-time.js (relative time module)
│ ├── relative-time/ (relative time source code)
│ ├── unit.js (unit module)
│ ├── unit/ (unit source code)
│ └── util/ (basic JavaScript helpers polyfills, eg array.map)
└── test/ (unit and functional test files)
├── fixtures/ (CLDR fixture data)
├── functional/ (functional tests)
├── functional.html
├── functional.js
├── unit/ (unit tests)
├── unit.html
└── unit.js
The source files are as granular as possible. When combined to generate the build file, all the excessive/overhead wrappers are cut off. It's following the same build model of jQuery and Modernizr.
Core, and all modules' public APIs are located in the src/
directory, ie. core.js
, date.js
, message.js
, number.js
, and plural.js
.
Install Grunt and external dependencies. First, install the grunt-cli and bower packages if you haven't before. These should be installed globally (like this: npm install -g grunt-cli bower
). Then:
npm install && bower install
Tests can be run either in the browser or using Node.js (via Grunt) after having installed the external development dependencies (for more details, see above).
To run the unit tests, run grunt test:unit
, or run grunt connect:keepalive
and open http://localhost:9001/test/unit.html
in a browser. It tests the very specific functionality of each function (sometimes internal/private).
The goal of the unit tests is to make it easy to spot bugs, easy to debug.
To run the functional tests, create the dist files by running grunt
. Then, run grunt test:functional
, or open http://localhost:9001/test/functional.html
in a browser. Note that grunt
will automatically run unit and functional tests for you to ensure the built files are safe.
The goal of the functional tests is to ensure that everything works as expected when it is combined.
Build the distribution files after having installed the external development dependencies (for more details, see above).
grunt