@sanvika/lang

Date: 23 April 2026 Centralized i18n SDK for the Sanvika ecosystem (50+ projects). Indian-12 / Global-28 presets, segment-based lazy loading, RTL, voice/TTS codes, server translations.

Table of Contents

1. Kya hai
2. Install
3. Quick Start (Indian 12)
4. Quick Start (Global 28)
5. API
6. Server Translations
7. Migration from raw i18next setup

Kya hai

Har Sanvika project mein i18n ka same boilerplate (200-700 lines, 192-560+ JSON imports) repeat ho raha tha. Yeh SDK us boilerplate ko single config call mein collapse karta hai. JSON files project mein hi rehti hain (project-owned), engine yahan se aata hai.

Install

pnpm add @sanvika/lang i18next react-i18next

Quick Start (Indian 12)

// src/utils/language/i18n.js
import { createSanvikaI18n, LANGUAGE_PRESETS } from "@sanvika/lang";

import enNavigation from "./translations/segments/en/navigation.json";
import hiNavigation from "./translations/segments/hi/navigation.json";
// ... other core segment imports

import enDashboard from "./translations/segments/en/dashboard.json";
import hiDashboard from "./translations/segments/hi/dashboard.json";
// ... other lazy segment imports

const { i18n, ensureTranslationSegment } = createSanvikaI18n({
  languages: LANGUAGE_PRESETS.INDIAN_12,
  defaultLanguage: "en",
  fallbackLanguage: "en",
  storageKey: "language",
  coreSegments: {
    en: { ...enNavigation /* , ...enUiCommon, ... */ },
    hi: { ...hiNavigation /* , ...hiUiCommon, ... */ },
  },
  lazySegments: {
    dashboard: { en: enDashboard, hi: hiDashboard },
  },
});

export default i18n;
export { ensureTranslationSegment };

Quick Start (Global 28)

import { createSanvikaI18n, LANGUAGE_PRESETS } from "@sanvika/lang";

const { i18n, ensureTranslationSegment } = createSanvikaI18n({
  languages: LANGUAGE_PRESETS.GLOBAL_28,
  defaultLanguage: "hi",
  fallbackLanguage: "en",
  coreSegments: { /* ... */ },
  lazySegments: { /* ... */ },
});

API

createSanvikaI18n(options)

Returns { i18n, ensureTranslationSegment, getCurrentLanguage }.

| Option | Type | Required | Default | |---|---|---|---| | languages | Array | yes | — | | coreSegments | Object | yes | — | | lazySegments | Object | no | {} | | defaultLanguage | string | no | "en" | | fallbackLanguage | string | no | "en" | | storageKey | string | no | "language" | | i18nextOptions | Object | no | {} |

<LanguageProvider validLanguages={[...]}>

Wraps your app, exposes useLanguage() hook with { currentLanguage, changeLanguage }. Auto-applies dir="rtl" for Urdu/Arabic.

Voice / Flag helpers

import {
  getVoiceLanguageCode,    // "hi" → "hi-IN"
  getFlagEmoji,             // "hi" → "🇮🇳"
  getNativeName,            // "ta" → "தமிழ்"
  isRTLLanguage,            // "ur" → true
  applyLanguageDirection,
  getSupportedLanguages,
} from "@sanvika/lang";

normalizeTranslationArray(value)

Coerces JSON list-like values (object → array, single → array, etc.) into proper arrays.

Server Translations

// src/utils/language/getTranslations.js
import { createServerTranslations } from "@sanvika/lang";
import enNavigation from "./translations/segments/en/navigation.json";
import hiNavigation from "./translations/segments/hi/navigation.json";

const { getTranslations, getServerLocale } = createServerTranslations({
  segmentDataMap: {
    navigation: { en: enNavigation, hi: hiNavigation },
  },
  validLocales: ["en", "hi", "ta", "te", "kn", "ml", "bn", "mr", "gu", "pa", "or", "ur"],
  fallbackLocale: "en",
});

export { getTranslations, getServerLocale };

Migration from raw i18next setup

| Before (per project) | After | |---|---| | 700-line i18n.js with 192/560+ static imports + manual resources, segmentLoaders, loadedSegments, ensureTranslationSegment | 1 createSanvikaI18n(...) call | | Hand-written voiceCodeMapper.js per project | Built-in (getVoiceLanguageCode, getFlagEmoji, etc.) | | Hand-written LanguageProvider/useLanguage per project | Built-in | | Hand-written getTranslations SSR helper per project | createServerTranslations(...) |