Switch branches/tags
Nothing to show
Find file Copy path
Fetching contributors…
Cannot retrieve contributors at this time
860 lines (814 sloc) 29.5 KB


A JavaScript library for globalization and localization. Enables complex culture-aware number and date parsing and formatting, including the raw culture information for hundreds of different languages and countries, as well as an extensible system for localization.

Why Globalization?

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.

What is a Culture?

Globalize defines roughly 350 cultures. Part of the reason for this large number, besides there being a lot of cultures in the world, is because for some languages, expectations differ among the countries that speak it. English, for example, is an official language in dozens of countries. Despite the language being English, the expected date formatting still greatly differs between them.

So, it does not seem useful to define cultures by their language alone. Nor is it useful to define a culture by its country alone, as many countries have several official languages, spoken by sizable populations. Therefore, cultures are defined as a combination of the language and the country speaking it. Each culture is given a unique code that is a combination of an ISO 639 two-letter lowercase culture code for the language, and a two-letter uppercase code for the country or region. For example, "en-US" is the culture code for English in the United States.

Yet, it is perhaps unreasonable to expect application developers to cater to every possible language/country combination perfectly. It is important then to define so-called "neutral" cultures based on each language. These cultures define the most likely accepted set of rules by anyone speaking that language, whatever the country. Neutral cultures are defined only by their language code. For example, "es" is the neutral culture for Spanish.

Globalize.addCultureInfo( cultureName, extendCultureName, info )

This method allows you to create a new culture based on an existing culture or add to existing culture info. If the optional argument

is not supplied, it will extend the existing culture if it exists or create a new culture based on the default culture if it doesn't exist. If cultureName is not supplied, it will add the supplied info to the current culture. See .culture().


A mapping of culture codes to culture objects. For example, is an object representing the complete culture definition for the neutral French culture. Note that the main globalize.js file alone only includes a neutral English culture. To get additional cultures, you must include one or more of the culture scripts that come with it. You can see in the section Defining Culture Information below which fields are defined in each culture.

Globalize.culture( selector )

An application that supports globalization and/or localization will need to have a way to determine the user's preference. Attempting to automatically determine the appropriate culture is useful, but it is good practice to always offer the user a choice, by whatever means.

Whatever your mechanism, it is likely that you will have to correlate the user's preferences with the list of cultures supported in the app. This method allows you to select the best match given the culture scripts that you have included and to set the Globalize culture to the culture which the user prefers.

If you pass an array of names instead of a single name string, the first culture for which there is a match (that culture's script has been referenced) will be used. If none match, the search restarts using the corresponding neutral cultures. For example, if the application has included only the neutral "fr" culture, any of these would select it:

Globalize.culture( "fr" );
console.log( Globalize.culture().name ) // "fr"

Globalize.culture( "fr-FR" ); console.log( Globalize.culture().name ) // "fr-FR"

Globalize.culture([ "es-MX", "fr-FR" ]); console.log( Globalize.culture().name ) // "es-MX"

In any case, if no match is found, the neutral English culture "en" is selected by default.

If you don't pass a selector, .culture() will return the current Globalize culture.

Each culture string may also follow the pattern defined in RFC2616 sec 14.4. That is, a culture name may include a "quality" value that indicates an estimate of the user's preference for the language.

Globalize.culture( "fr;q=0.4, es;q=0.5, he" );

In this example, the neutral Hebrew culture "he" is given top priority (an unspecified quality is equal to 1). If that language is not an exact match for any of the cultures available in Globalize.cultures, then "es" is the next highest priority with 0.5, etc. If none of these match, just like with the array syntax, the search starts over and the same rules are applied to the corresponding neutral language culture for each. If still none match, the neutral English culture "en" is used.

Globalize.findClosestCulture( selector )

Just like .culture( selector ), but it just returns the matching culture, if any, without setting it to the current Globalize culture, returned by .culture().

Globalize.format( value, format, culture )

Formats a date or number according to the given format string and the given culture (or the current culture if not specified). See the sections Number Formatting and Date Formatting below for details on the available formats.

// assuming a culture with number grouping of 3 digits,
// using "," separator and "." decimal symbol.
Globalize.format( 1234.567, "n" ); // "1,234.57"
Globalize.format( 1234.567, "n1" ); // "1,234.6"
Globalize.format( 1234.567, "n0" ); // "1,235"

// assuming a culture with "/" as the date separator symbol Globalize.format( new Date(1955,10,5), "yyyy/MM/dd" ); // "1955/11/05" Globalize.format( new Date(1955,10,5), "dddd MMMM d, yyyy" ); // "Saturday November 5, 1955"

Globalize.localize( key, culture )

Gets or sets a localized value. This method allows you to extend the information available to a particular culture, and to easily retrieve it without worrying about finding the most appropriate culture. For example, to define the word "translate" in French:

Globalize.addCultureInfo( "fr", {
	messages: {
		"translate": "traduire"
console.log( Globalize.localize( "translate", "fr" ) ); // "traduire"
Note that localize() will find the closest match available per the same semantics as the Globalize.findClosestCulture() method. If there is no match, the translation given is for the neutral English culture "en" by default.

Globalize.parseInt( value, radix, culture )

Parses a string representing a whole number in the given radix (10 by default), taking into account any formatting rules followed by the given culture (or the current culture, if not specified).

// assuming a culture where "," is the group separator
// and "." is the decimal separator
Globalize.parseInt( "1,234.56" ); // 1234
// assuming a culture where "." is the group separator
// and "," is the decimal separator
Globalize.parseInt( "1.234,56" ); // 1234

Globalize.parseFloat( value, radix, culture )

Parses a string representing a floating point number in the given radix (10 by default), taking into account any formatting rules followed by the given culture (or the current culture, if not specified).

// assuming a culture where "," is the group separator
// and "." is the decimal separator
Globalize.parseFloat( "1,234.56" ); // 1234.56
// assuming a culture where "." is the group separator
// and "," is the decimal separator
Globalize.parseFloat( "1.234,56" ); // 1234.56

Globalize.parseDate( value, formats, culture )

Parses a string representing a date into a JavaScript Date object, taking into account the given possible formats (or the given culture's set of default formats if not given). As before, the current culture is used if one is not specified.

Globalize.culture( "en" );
Globalize.parseDate( "1/2/2003" ); // Thu Jan 02 2003
Globalize.culture( "fr" );
Globalize.parseDate( "1/2/2003" ); // Sat Feb 01 2003

Utilizing and Extending Cultures

The culture information included with each culture is mostly necessary for the parsing and formatting methods, but not all of it. For example, the Native and English names for each culture is given, as well as a boolean indicating whether the language is right-to-left. This may be useful information for your own purposes. You may also add to the culture information directly if so desired.

As an example, in the U.S., the word "billion" means the number 1,000,000,000 (9 zeros). But in other countries, that number is "1000 million" or a "milliard", and a billion is 1,000,000,000,000 (12 zeros). If you needed to provide functionality to your app or custom plugin that needed to know how many zeros are in a "billion", you could extend the culture information as follows:

// define additional culture information for a possibly existing culture
Globalize.addCultureInfo( "fr", {
	numberFormat: {
		billionZeroes: 12
Using this mechanism, the "fr" culture will be created if it does not exist. And if it does, the given values will be added to it.

Defining Culture Information

Each culture is defined in its own script with the naming scheme globalize.culture.<name>.js. You may include any number of these scripts, making them available in the Globalize.cultures mapping. Including one of these scripts does NOT automatically make it the current culture selected in the Globalize.culture property.

The neutral English culture is defined directly in globalize.js, and set both to the properties "en" and "default" of the Globalize.cultures mapping. Extensive comments describe the purpose of each of the fields defined.

Looking at the source code of the scripts for each culture, you will notice that each script uses Globalize.addCultureInfo() to have the "default" neutral English culture "en", as a common basis, and defines only the properties that differ from neutral English.

The neutral English culture is listed here along with the comments:

Globalize.cultures[ "default" ] = {
	// A unique name for the culture in the form
	// <language code>-<country/region code>
	name: "English",
	// the name of the culture in the English language
	englishName: "English",
	// the name of the culture in its own language
	nativeName: "English",
	// whether the culture uses right-to-left text
	isRTL: false,
	// "language" is used for so-called "specific" cultures.
	// For example, the culture "es-CL" means Spanish in Chili.
	// It represents the Spanish-speaking culture as it is in Chili,
	// which might have different formatting rules or even translations
	// than Spanish in Spain. A "neutral" culture is one that is not
	// specific to a region. For example, the culture "es" is the generic
	// Spanish culture, which may be a more generalized version of the language
	// that may or may not be what a specific culture expects.
	// For a specific culture like "es-CL", the "language" field refers to the
	// neutral, generic culture information for the language it is using.
	// This is not always a simple matter of the string before the dash.
	// For example, the "zh-Hans" culture is neutral (Simplified Chinese).
	// And the "zh-SG" culture is Simplified Chinese in Singapore, whose
	// language field is "zh-CHS", not "zh".
	// This field should be used to navigate from a specific culture to its
	// more general, neutral culture. If a culture is already as general as it
	// can get, the language may refer to itself.
	language: "en",
	// "numberFormat" defines general number formatting rules, like the digits
	// in each grouping, the group separator, and how negative numbers are
	// displayed.
	numberFormat: {
		// [negativePattern]
		// Note, numberFormat.pattern has no "positivePattern" unlike percent
		// and currency, but is still defined as an array for consistency with
		// them.
		//	  negativePattern: one of "(n)|-n|- n|n-|n -"
		pattern: [ "-n" ],
		// number of decimal places normally shown
		decimals: 2,
		// string that separates number groups, as in 1,000,000
		",": ",",
		// string that separates a number from the fractional portion,
		// as in 1.99
		".": ".",
		// array of numbers indicating the size of each number group.
		groupSizes: [ 3 ],
		// symbol used for positive numbers
		"+": "+",
		// symbol used for negative numbers
		"-": "-",
		percent: {
			// [negativePattern, positivePattern]
			//	   negativePattern: one of "-n %|-n%|-%n|%-n|%n-|n-%|n%-|-% n|n %-|% n-|% -n|n- %"
			//	   positivePattern: one of "n %|n%|%n|% n"
			pattern: [ "-n %", "n %" ],
			// number of decimal places normally shown
			decimals: 2,
			// array of numbers indicating the size of each number group.
			groupSizes: [ 3 ],
			// string that separates number groups, as in 1,000,000
			",": ",",
			// string that separates a number from the fractional portion, as in 1.99
			".": ".",
			// symbol used to represent a percentage
			symbol: "%"
		currency: {
			// [negativePattern, positivePattern]
			//	   negativePattern: one of "($n)|-$n|$-n|$n-|(n$)|-n$|n-$|n$-|-n $|-$ n|n $-|$ n-|$ -n|n- $|($ n)|(n $)"
			//	   positivePattern: one of "$n|n$|$ n|n $"
			pattern: [ "($n)", "$n" ],
			// number of decimal places normally shown
			decimals: 2,
			// array of numbers indicating the size of each number group.
			groupSizes: [ 3 ],
			// string that separates number groups, as in 1,000,000
			",": ",",
			// string that separates a number from the fractional portion, as in 1.99
			".": ".",
			// symbol used to represent currency
			symbol: "$"
	// "calendars" property defines all the possible calendars used by this
	// culture. There should be at least one defined with name "standard" which
	// is the default calendar used by the culture.
	// A calendar contains information about how dates are formatted,
	// information about the calendar's eras, a standard set of the date
	// formats, translations for day and month names, and if the calendar is
	// not based on the Gregorian calendar, conversion functions to and from
	// the Gregorian calendar.
	calendars: {
		standard: {
			// name that identifies the type of calendar this is
			name: "Gregorian_USEnglish",
			// separator of parts of a date (e.g. "/" in 11/05/1955)
			"/": "/",
			// separator of parts of a time (e.g. ":" in 05:44 PM)
			":": ":",
			// the first day of the week (0 = Sunday, 1 = Monday, etc)
			firstDay: 0,
			days: {
				// full day names
				names: [ "Sunday", "Monday", "Tuesday", "Wednesday", "Thursday", "Friday", "Saturday" ],
				// abbreviated day names
				namesAbbr: [ "Sun", "Mon", "Tue", "Wed", "Thu", "Fri", "Sat" ],
				// shortest day names
				namesShort: [ "Su", "Mo", "Tu", "We", "Th", "Fr", "Sa" ]
			months: [
				// full month names (13 months for lunar calendars -- 13th month should be "" if not lunar)
				names: [ "January", "February", "March", "April", "May", "June", "July", "August", "September", "October", "November", "December", "" ],
				// abbreviated month names
				namesAbbr: [ "Jan", "Feb", "Mar", "Apr", "May", "Jun", "Jul", "Aug", "Sep", "Oct", "Nov", "Dec", "" ]
			// AM and PM designators in one of these forms:
			// The usual view, and the upper and lower case versions
			//		[standard,lowercase,uppercase]
			// The culture does not use AM or PM (likely all standard date
			// formats use 24 hour time)
			//		null
			AM: [ "AM", "am", "AM" ],
			PM: [ "PM", "pm", "PM" ],
			eras: [
				// eras in reverse chronological order.
				// name: the name of the era in this culture (e.g. A.D., C.E.)
				// start: when the era starts in ticks, null if it is the
				//		  earliest supported era.
				// offset: offset in years from gregorian calendar
			// when a two digit year is given, it will never be parsed as a
			// four digit year greater than this year (in the appropriate era
			// for the culture)
			// Set it as a full year (e.g. 2029) or use an offset format
			// starting from the current year: "+19" would correspond to 2029
			// if the current year is 2010.
			twoDigitYearMax: 2029,
			// set of predefined date and time patterns used by the culture.
			// These represent the format someone in this culture would expect
			// to see given the portions of the date that are shown.
			patterns: {
				// short date pattern
				d: "M/d/yyyy",
				// long date pattern
				D: "dddd, MMMM dd, yyyy",
				// short time pattern
				t: "h:mm tt",
				// long time pattern
				T: "h:mm:ss tt",
				// long date, short time pattern
				f: "dddd, MMMM dd, yyyy h:mm tt",
				// long date, long time pattern
				F: "dddd, MMMM dd, yyyy h:mm:ss tt",
				// month/day pattern
				M: "MMMM dd",
				// month/year pattern
				Y: "yyyy MMMM",
				// S is a sortable format that does not vary by culture
				S: "yyyy\u0027-\u0027MM\u0027-\u0027dd\u0027T\u0027HH\u0027:\u0027mm\u0027:\u0027ss"
			// optional fields for each calendar:
				Same as months but used when the day preceeds the month.
				Omit if the culture has no genitive distinction in month names.
				For an explanation of genitive months, see
				Allows for the support of non-gregorian based calendars. This
				"convert" object defines two functions to convert a date to and
				from a gregorian calendar date:
					fromGregorian( date )
						Given the date as a parameter, return an array with
						parts [ year, month, day ] corresponding to the
						non-gregorian based year, month, and day for the
					toGregorian( year, month, day )
						Given the non-gregorian year, month, and day, return a
						new Date() object set to the corresponding date in the
						gregorian calendar.
	// Map of messages used by .localize()
	messages: {}

Each culture can have several possible calendars. The calendar named "standard" is the default calendar used by that culture. You may change the calendar in use by setting the "calendar" field. Take a look at the calendars defined by each culture by looking at the script or enumerating its calendars collection.

// switch to a non-standard calendar
Globalize.culture().calendar = Globalize.culture().calendars.SomeOtherCalendar;
// back to the standard calendar
Globalize.culture().calendar = Globalize.culture().calendars.standard;

Number Formatting

When formatting a number with format(), the main purpose is to convert the number into a human readable string using the culture's standard grouping and decimal rules. The rules between cultures can vary a lot. For example, in some cultures, the grouping of numbers is done unevenly. In the "te-IN" culture (Telugu in India), groups have 3 digits and then 2 digits. The number 1000000 (one million) is written as "10,00,000". Some cultures do not group numbers at all.

There are four main types of number formatting:

  • n for number
  • d for decimal digits
  • p for percentage
  • c for currency
Even within the same culture, the formatting rules can vary between these four types of numbers. For example, the expected number of decimal places may differ from the number format to the currency format. Each format token may also be followed by a number. The number determines how many decimal places to display for all the format types except decimal, for which it means the minimum number of digits to display, zero padding it if necessary. Also note that the way negative numbers are represented in each culture can vary, such as what the negative sign is, and whether the negative sign appears before or after the number. This is especially apparent with currency formatting, where many cultures use parentheses instead of a negative sign.
// just for example - will vary by culture
Globalize.format( 123.45, "n" ); // 123.45
Globalize.format( 123.45, "n0" ); // 123
Globalize.format( 123.45, "n1" ); // 123.5

Globalize.format( 123.45, "d" ); // 123 Globalize.format( 12, "d3" ); // 012

Globalize.format( 123.45, "c" ); // $123.45 Globalize.format( 123.45, "c0" ); // $123 Globalize.format( 123.45, "c1" ); // $123.5 Globalize.format( -123.45, "c" ); // ($123.45)

Globalize.format( 0.12345, "p" ); // 12.35 % Globalize.format( 0.12345, "p0" ); // 12 % Globalize.format( 0.12345, "p4" ); // 12.3450 %

Parsing with parseInt and parseFloat also accepts any of these formats.

Currency Formatting

Globalize has a default currency symbol for each locale. This is used when formatting a currency value such as

Globalize.format( 1234.56, "c" ); // $1,234.56
You can change the currency symbol for a locale by modifying the culture's numberFormat.currency.symbol property:
Globalize.culture( "en-US" ).numberFormat.currency.symbol = '\u20ac'; // euro sign U+20AC
If you need to switch between currency symbols, you could write a function to do that, such as
function setCurrency( currSym ) {
  Globalize.culture().numberFormat.currency.symbol = currSym;

Date Formatting

Date formatting varies wildly by culture, not just in the spelling of month and day names, and the date separator, but by the expected order of the various date components, whether to use a 12 or 24 hour clock, and how months and days are abbreviated. Many cultures even include "genitive" month names, which are different from the typical names and are used only in certain cases.

Also, each culture has a set of "standard" or "typical" formats. For example, in "en-US", when displaying a date in its fullest form, it looks like "Saturday, November 05, 1955". Note the non-abbreviated day and month name, the zero padded date, and four digit year. So, Globalize expects a certain set of "standard" formatting strings for dates in the "patterns" property of the "standard" calendar of each culture, that describe specific formats for the culture. The third column shows example values in the neutral English culture "en-US"; see the second table for the meaning tokens used in date formats.

// just for example - will vary by culture
Globalize.format( new Date(2012, 1, 20), 'd' ); // 2/20/2012
Globalize.format( new Date(2012, 1, 20), 'D' ); // Monday, February 20, 2012

Format Meaning "en-US"
f Long Date, Short Time dddd, MMMM dd, yyyy h:mm tt
F Long Date, Long Time dddd, MMMM dd, yyyy h:mm:ss tt
t Short Time h:mm tt
T Long Time h:mm:ss tt
d Short Date M/d/yyyy
D Long Date dddd, MMMM dd, yyyy
Y Month/Year MMMM, yyyy
M Month/Day MMMM dd

In addition to these standard formats, there is the "S" format. This is a sortable format that is identical in every culture: "yyyy'-'MM'-'dd'T'HH':'mm':'ss".

When more specific control is needed over the formatting, you may use any format you wish by specifying the following custom tokens:

Token Meaning Example
d Day of month (no leading zero) 5
dd Day of month (leading zero) 05
ddd Day name (abbreviated) Sat
dddd Day name (full) Saturday
M Month of year (no leading zero) 9
MM Month of year (leading zero) 09
MMM Month name (abbreviated) Sep
MMMM Month name (full) September
yy Year (two digits) 55
yyyy Year (four digits) 1955
'literal' Literal Text 'of the clock'
\' Single Quote 'o'\''clock'
m Minutes (no leading zero) 9
mm Minutes (leading zero) 09
h Hours (12 hour time, no leading zero) 6
hh Hours (12 hour time, leading zero) 06
H Hours (24 hour time, no leading zero) 5 (5am) 15 (3pm)
HH Hours (24 hour time, leading zero) 05 (5am) 15 (3pm)
s Seconds (no leading zero) 9
ss Seconds (leading zero) 09
f Deciseconds 1
ff Centiseconds 11
fff Milliseconds 111
t AM/PM indicator (first letter) A or P
tt AM/PM indicator (full) AM or PM
z Timezone offset (hours only, no leading zero) -8
zz Timezone offset (hours only, leading zero) -08
zzz Timezone offset (full hours/minutes) -08:00
g or gg Era name A.D.

Generating Culture Files

The Globalize culture files are generated JavaScript containing metadata and functions based on culture info in the Microsoft .Net Framework 4.


Building the generator

  1. Open a Windows Command Prompt ( Start -> Run... -> cmd )
  2. Change directory to root of Globalize project (where file is located)
  3. "C:\Windows\Microsoft.NET\Framework\v4.0.30319\MSBuild" generator\generator.csproj

Running the generator

  1. Open a Windows Command Prompt
  2. Change directory to root of Globalize project (where file is located)
  3. "generator\bin\Debug\generator.exe"

Building Globalize

Globalize is built using grunt, a node-based build utility. First, make sure grunt is installed globally:

> npm install -g grunt

then you can lint and test by simply running grunt in the globalize folder

> cd globalize
> grunt