JSPM – i18n-js@4.0.0-alpha.6

Package Exports

i18n-js

Readme

i18n-js

A small library to provide the i18n translations on the JavaScript.

Installation

Yarn: yarn add i18n-js@next
NPM: npm install i18n-js@next

Usage

Setting up

First, you need to instantiate I18n with the translations' object, the main class of this library.

import { I18n } from "i18n";
import translations from "./translations.json";

const i18n = new I18n(translations);

The translations object is a direct export of translations defined by Ruby on Rails. To export the translations, you can use i18n-js, a Ruby gem that's completely disconnected from Rails and that can be used for the solely purpose of exporting the translations, even if your project is written in a different language. If all you care about is some basic translation mechanism, then you can set the object like this:

const i18n = new I18n({
  en: {
    hello: "Hi!",
  },
  "pt-BR": {
    hello: "Olá!",
  },
});

Each root key is a different locale that may or may not have the script code. This library also supports locales with region code, like zh-Hant-TW.

Once everything is set up, you can then define the locale. en is both the current and default locale. To override either values, you have to use I18n#defaultLocale and I18n#locale.

i18n.defaultLocale = "pt-BR";
i18n.locale = "pt-BR";

Base translations

This library comes bundled with all base translations made available by rails-i18n. Base translations allow formatting date, numbers, and sentence connectors, among other things.

To load the base translations, use something like the following:

import { I18n } from "i18n-js";
import ptBR from "i18n-js/json/pt-BR.json";
import en from "i18n-js/json/en.json";

const i18n = new I18n({
  ...ptBR,
  ...en,
});

Updating translation store

Updating the translation store is trivial. All you have to do is calling I18n#store with the translations that need to be merged. Let's assume you've exported all your app's translations using i18n-js CLI, using a separate file for each language, like this:

translations/en.json
translations/pt-BR.json

This is how you could update the store:

import { I18n } from "i18n-js";
import ptBR from "translations/pt-BR.json";
import en from "translations/en.json";

const i18n = new I18n();

i18n.store(en);
i18n.store(ptBR);

This method will allow you to lazy load translations and them updating the store as needed.

import { I18n } from "i18n-js";

async function loadTranslations(i18n, locale) {
  const response = await fetch(`/translations/${locale}.json`);
  const translations = await response.json();

  i18n.store(translations);
}

const i18n = new I18n();
loadTranslations(i18n, "es");

Translating messages

To translate messages, you have to use the I18n#translate, or its I18n#t alias.

i18n.locale = "en";
i18n.t("hello"); //=> Hi!

i18n.locale = "pt-BR";
i18n.t("hello"); //=> Olá!

You can also provide an array as scope. Both calls below are equivalent.

i18n.t(["greetings", "hello"]);
i18n.t("greetings.hello");

Your translations may have dynamic values that should be interpolated. Here's a greeting message that takes a name:

const i18n = new I18n({
  en: { greetings: "Hi, %{name}!" },
  "pt-BR": { greetings: "Olá, %{name}!" },
});

i18n.t("greetings", { name: "John" });

Missing translations

A translation may be missing. In that case, you may set the default value that's going to be returned.

i18n.t("missing.scope", { defaultValue: "This is a default message" });

Default messages can also have interpolation.

i18n.t("noun", { defaultValue: "I'm a {{noun}}", noun: "Mac" });

Alternatively, you can define a list of scopes that will be searched instead.

// As a scope
i18n.t("some.missing.scope", { defaults: [{ scope: "some.existing.scope" }] });

// As a simple translation
i18n.t("some.missing.scope", { defaults: [{ message: "Some message" }] });

Default values must be provided as an array of objects where the key is the type of desired translation, a scope or a message. The returned translation will be either the first scope recognized, or the first message defined.

The translation will fall back to the defaultValue translation if no scope in defaults matches and if no message default is found.

You can enable translation fallback with I18n#enableFallback.

i18n.enableFallback = true;

By default missing translations will first be looked for in less specific versions of the requested locale and if that fails by taking them from your I18n#defaultLocale.

// if i18n.defaultLocale = "en" and translation doesn't exist
// for i18n.locale = "de-DE" this key will be taken from "de" locale scope
// or, if that also doesn't exist, from "en" locale scope
i18n.t("some.missing.scope");

Custom fallback rules can also be specified for a specific language. There are three different ways of doing it so. In any case, the locale handler must be registered using i18n.locales.register().

// Using an array
i18n.locales.register("no", ["nb", "en"]);

// Using a string
i18n.locales.no.register("nb");

// Using a function.
i18n.locales.no.register((locale) => ["nb"]);

By default a missing translation will be displayed as [missing "name of scope" translation]. You can override this behavior by setting i18n.missingBehavior to "guess".

i18n.missingBehavior = "guess";

The "guess" behavior will take the last section of the scope and apply some replace rules; camel case becomes lower case and underscores are replaced with space. In practice, it means that a scope like questionnaire.whatIsYourFavorite_ChristmasPresent becomes what is your favorite Christmas present.

There's also a strategy called error, which will throw an exception every time you fetch a missing translation. This is great for development. It'll even end up on your error tracking!

i18n.missingBehavior = "error";

To detect missing translations, you can also set i18n.missingTranslationPrefix.

i18n.missingTranslationPrefix = "EE: ";

The same questionnaire.whatIsYourFavorite_ChristmasPresent scope would converted into EE: what is your favorite Christmas present. This is helpful if you want to add a check to your automated tests.

Finally, you can completely override the missing translation strategy by setting it to a function. The following example will return null for every missing translation.

i18n.missingTranslation = () => null;

Pluralization

This library has support for pluralization and by default works with English, and similar pluralized languages like Portuguese.

First, you have to define your translations with special keywords defined by the pluralization handler. The default keywords are zero, one, and other.

const i18n = new I18n({
  en: {
    inbox: {
      zero: "You have no messages",
      one: "You have one message",
      other: "You have %{count} messages",
    },
  },

  "pt-BR": {
    inbox: {
      zero: "Você não tem mensagens",
      one: "Você tem uma mensagem",
      other: "Você tem %{count} mensagens",
    },
  },
});

To retrieve the pluralized translation you must provide the count option with a numeric value.

i18n.t("inbox", { count: 0 }); //=> You have no messages
i18n.t("inbox", { count: 1 }); //=> You have one message
i18n.t("inbox", { count: 2 }); //=> You have 2 messages

You may need to define new rules for other languages like Russian. This can be done by registering a handler with i18n.pluralization.register(). The following example defines a Russian pluralizer.

i18n.pluralization.register("ru", (_i18n, count) => {
  const mod10 = count % 10;
  const mod100 = count % 100;
  let key;

  const one = mod10 === 1 && mod100 !== 11;
  const few = [2, 3, 4].includes(mod10) && ![12, 13, 14].includes(mod100);
  const many =
    mod10 === 0 ||
    [5, 6, 7, 8, 9].includes(mod10) ||
    [11, 12, 13, 14].includes(mod100);

  if (one) {
    key = "one";
  } else if (few) {
    key = "few";
  } else if (many) {
    key = "many";
  } else {
    key = "other";
  }

  return [key];
});

You can find all rules on http://www.unicode.org/.

It's encourage to publish your pluralizers using the following name pattern: i18n-<locale>-pluralizer. If you publish a pluralizer, please add a pull request so we can list it here.

Other options

If you're providing the same scope again and again, you can reduce the boilerplate by setting the scope option.

const options = { scope: "activerecord.attributes.user" };

i18n.t("name", options);
i18n.t("email", options);
i18n.t("username", options);

Number Formatting

Similar to Rails helpers, you can have localized number and currency formatting.

i18n.l("currency", 1990.99);
// $1,990.99

i18n.l("number", 1990.99);
// 1,990.99

i18n.l("percentage", 123.45);
// 123.450%

To have more control over number formatting, you can use the I18n#toNumber, I18n#toPercentage, I18n#toCurrency and I18n#toHumanSize functions.

i18n.toNumber(1000); // 1,000.000
i18n.toCurrency(1000); // $1,000.00
i18n.toPercentage(100); // 100.000%

The I18n#toNumber and I18n#toPercentage functions accept the following options:

precision: defaults to 3
separator: defaults to .
delimiter: defaults to ,
stripInsignificantZeros: defaults to false

See some number formatting examples:

i18n.toNumber(1000, { precision: 0 }); // 1,000
i18n.toNumber(1000, { delimiter: ".", separator: "," }); // 1.000,000
i18n.toNumber(1000, { delimiter: ".", precision: 0 }); // 1.000

The I18n#toCurrency function accepts the following options:

precision: sets the level of precision
separator: sets the separator between the units
delimiter: sets the thousands delimiter
format: sets the format of the output string
unit: sets the denomination of the currency
stripInsignificantZeros: defaults to false
signFirst: defaults to true

You can provide only the options you want to override:

i18n.toCurrency(1000, { precision: 0 }); // $1,000

The I18n#toHumanSize function accepts the following options:

precision: defaults to 1
separator: defaults to .
delimiter: defaults to ""
stripInsignificantZeros: defaults to false
format: defaults to %n%u

i18n.toHumanSize(1234); // 1KB
i18n.toHumanSize(1234 * 1024); // 1MB

Date Formatting

The I18n#localize (or its alias I18n#l) can accept a string, epoch time integer or a Date object. You can see below the accepted formats:

// yyyy-mm-dd
i18n.l("date.formats.short", "2009-09-18");

// yyyy-mm-dd hh:mm:ss
i18n.l("time.formats.short", "2009-09-18 23:12:43");

// JSON format with local Timezone (part of ISO-8601)
i18n.l("time.formats.short", "2009-11-09T18:10:34");

// JSON format in UTC (part of ISO-8601)
i18n.l("time.formats.short", "2009-11-09T18:10:34Z");

// Epoch time
i18n.l("date.formats.short", 1251862029000);

// mm/dd/yyyy
i18n.l("date.formats.short", "09/18/2009");

// Date object
i18n.l("date.formats.short", new Date());

You can also add placeholders to the date format:

const i18n = new I18n({
  date: {
    formats: {
      ordinalDay: "%B %{day}",
    },
  },
});

i18n.l("date.formats.ordinalDay", "2009-09-18", { day: "18th" }); // Sep 18th

If you prefer, you can use the I18n#toTime and I18n#strftime functions directly to format dates.

var date = new Date();
i18n.toTime("date.formats.short", "2009-09-18");
i18n.toTime("date.formats.short", date);
i18n.strftime(date, "%d/%m/%Y");

The accepted formats for i18n.strftime are:

%a  - The abbreviated weekday name (Sun)
%A  - The full weekday name (Sunday)
%b  - The abbreviated month name (Jan)
%B  - The full month name (January)
%c  - The preferred local date and time representation
%d  - Day of the month (01..31)
%-d - Day of the month (1..31)
%H  - Hour of the day, 24-hour clock (00..23)
%-H - Hour of the day, 24-hour clock (0..23)
%k  - Hour of the day, 24-hour clock (0..23)
%I  - Hour of the day, 12-hour clock (01..12)
%-I - Hour of the day, 12-hour clock (1..12)
%l  - Hour of the day, 12-hour clock (1..12)
%m  - Month of the year (01..12)
%-m - Month of the year (1..12)
%M  - Minute of the hour (00..59)
%-M - Minute of the hour (0..59)
%p  - Meridian indicator (AM  or  PM)
%P  - Meridian indicator (am  or  pm)
%S  - Second of the minute (00..60)
%-S - Second of the minute (0..60)
%w  - Day of the week (Sunday is 0, 0..6)
%y  - Year without a century (00..99)
%-y - Year without a century (0..99)
%Y  - Year with century
%z  - Timezone offset (+0545)
%Z  - Timezone offset (+0545)

Check out __tests__/strftime.test.ts file for more examples!

Finally, you can also diplay relative time strings using I18n#timeAgoInWords.

const to = new Date();
const from = to.getTime() - 60 * 60 * 1000; // ~1h ago.

i18n.timeAgoInWords(from, to);
//=> about 1 hour

Using pluralization and number formatting together

Sometimes you might want to display translation with formatted number, like adding thousand delimiters to displayed number You can do this:

const i18n = new I18n({
  en: {
    points: {
      one: "1 Point",
      other: "{{points}} Points",
    },
  },
});

const points = 1234;

i18n.t("points", {
  count: points,
  points: i18n.toNumber(points),
});

Output should be 1,234 points.

Other helpers

`I18n#toSentence(list, options)`

i18n.toSentence(["apple", "banana", "pineapple"]);
//=> apple, banana, and pineapple.

License

(The MIT License)

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the 'Software'), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED 'AS IS', WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.