@lokascript/framework 7.0.0 → 7.1.0

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@lokascript/framework",
-  "version": "7.0.0",
+  "version": "7.1.0",
   "description": "Generic framework for building multilingual DSLs with semantic parsing and grammar transformation",
   "type": "module",
   "main": "dist/index.cjs",
@@ -28,6 +28,131 @@
       "types": "./dist/multilingual/index.d.ts",
       "import": "./dist/multilingual/index.js"
     },
+    "./languages": {
+      "types": "./dist/languages/index.d.ts",
+      "import": "./dist/languages/index.js",
+      "require": "./dist/languages/index.cjs"
+    },
+    "./languages/ar": {
+      "types": "./dist/languages/ar.d.ts",
+      "import": "./dist/languages/ar.js",
+      "require": "./dist/languages/ar.cjs"
+    },
+    "./languages/bn": {
+      "types": "./dist/languages/bn.d.ts",
+      "import": "./dist/languages/bn.js",
+      "require": "./dist/languages/bn.cjs"
+    },
+    "./languages/de": {
+      "types": "./dist/languages/de.d.ts",
+      "import": "./dist/languages/de.js",
+      "require": "./dist/languages/de.cjs"
+    },
+    "./languages/en": {
+      "types": "./dist/languages/en.d.ts",
+      "import": "./dist/languages/en.js",
+      "require": "./dist/languages/en.cjs"
+    },
+    "./languages/es": {
+      "types": "./dist/languages/es.d.ts",
+      "import": "./dist/languages/es.js",
+      "require": "./dist/languages/es.cjs"
+    },
+    "./languages/fr": {
+      "types": "./dist/languages/fr.d.ts",
+      "import": "./dist/languages/fr.js",
+      "require": "./dist/languages/fr.cjs"
+    },
+    "./languages/he": {
+      "types": "./dist/languages/he.d.ts",
+      "import": "./dist/languages/he.js",
+      "require": "./dist/languages/he.cjs"
+    },
+    "./languages/hi": {
+      "types": "./dist/languages/hi.d.ts",
+      "import": "./dist/languages/hi.js",
+      "require": "./dist/languages/hi.cjs"
+    },
+    "./languages/id": {
+      "types": "./dist/languages/id.d.ts",
+      "import": "./dist/languages/id.js",
+      "require": "./dist/languages/id.cjs"
+    },
+    "./languages/it": {
+      "types": "./dist/languages/it.d.ts",
+      "import": "./dist/languages/it.js",
+      "require": "./dist/languages/it.cjs"
+    },
+    "./languages/ja": {
+      "types": "./dist/languages/ja.d.ts",
+      "import": "./dist/languages/ja.js",
+      "require": "./dist/languages/ja.cjs"
+    },
+    "./languages/ko": {
+      "types": "./dist/languages/ko.d.ts",
+      "import": "./dist/languages/ko.js",
+      "require": "./dist/languages/ko.cjs"
+    },
+    "./languages/ms": {
+      "types": "./dist/languages/ms.d.ts",
+      "import": "./dist/languages/ms.js",
+      "require": "./dist/languages/ms.cjs"
+    },
+    "./languages/pl": {
+      "types": "./dist/languages/pl.d.ts",
+      "import": "./dist/languages/pl.js",
+      "require": "./dist/languages/pl.cjs"
+    },
+    "./languages/pt": {
+      "types": "./dist/languages/pt.d.ts",
+      "import": "./dist/languages/pt.js",
+      "require": "./dist/languages/pt.cjs"
+    },
+    "./languages/qu": {
+      "types": "./dist/languages/qu.d.ts",
+      "import": "./dist/languages/qu.js",
+      "require": "./dist/languages/qu.cjs"
+    },
+    "./languages/ru": {
+      "types": "./dist/languages/ru.d.ts",
+      "import": "./dist/languages/ru.js",
+      "require": "./dist/languages/ru.cjs"
+    },
+    "./languages/sw": {
+      "types": "./dist/languages/sw.d.ts",
+      "import": "./dist/languages/sw.js",
+      "require": "./dist/languages/sw.cjs"
+    },
+    "./languages/th": {
+      "types": "./dist/languages/th.d.ts",
+      "import": "./dist/languages/th.js",
+      "require": "./dist/languages/th.cjs"
+    },
+    "./languages/tl": {
+      "types": "./dist/languages/tl.d.ts",
+      "import": "./dist/languages/tl.js",
+      "require": "./dist/languages/tl.cjs"
+    },
+    "./languages/tr": {
+      "types": "./dist/languages/tr.d.ts",
+      "import": "./dist/languages/tr.js",
+      "require": "./dist/languages/tr.cjs"
+    },
+    "./languages/uk": {
+      "types": "./dist/languages/uk.d.ts",
+      "import": "./dist/languages/uk.js",
+      "require": "./dist/languages/uk.cjs"
+    },
+    "./languages/vi": {
+      "types": "./dist/languages/vi.d.ts",
+      "import": "./dist/languages/vi.js",
+      "require": "./dist/languages/vi.cjs"
+    },
+    "./languages/zh": {
+      "types": "./dist/languages/zh.d.ts",
+      "import": "./dist/languages/zh.js",
+      "require": "./dist/languages/zh.cjs"
+    },
     "./schema": {
       "types": "./dist/schema/index.d.ts",
       "import": "./dist/schema/index.js"
@@ -95,15 +220,15 @@
   "author": "LokaScript Contributors",
   "license": "MIT",
   "dependencies": {
-    "@lokascript/intent": "^7.0.0"
+    "@lokascript/intent": "^7.1.0"
   },
   "devDependencies": {
     "@lokascript/domains": "^5.0.0",
-    "@types/node": "^26.6.3",
-    "@vitest/coverage-v8": "^5.0.0",
+    "@types/node": "^26.6.5",
+    "@vitest/coverage-v8": "^5.0.5",
     "tsup": "^9.0.0",
     "typescript": "^5.0.0",
-    "vitest": "^5.0.2"
+    "vitest": "^5.0.5"
   },
   "files": [
     "dist",

package/src/index.ts

@@ -65,6 +65,8 @@ export type {
   DomainTokenizerOptions,
   LanguageConfigMeta,
 } from './multilingual';
+// A language's grammar (the data lives in `@lokascript/framework/languages/<code>`)
+export type { GrammarProfile } from './languages/types';
 
 // Core - export classes and utilities
 export * from './core';

package/src/languages/ar.ts

@@ -0,0 +1,41 @@
+/**
+ * Arabic Grammar
+ *
+ * VSO word order, prepositions, RTL (right-to-left), space-separated.
+ * Features root-based morphology and rich verb conjugation.
+ */
+
+import type { GrammarProfile } from './types';
+
+export const arabicGrammar = {
+  code: 'ar',
+  name: 'Arabic',
+  nativeName: 'العربية',
+  direction: 'rtl',
+  script: 'arabic',
+  wordOrder: 'VSO',
+  markingStrategy: 'preposition',
+  usesSpaces: true,
+  verb: {
+    position: 'start',
+    subjectDrop: true,
+  },
+  possessive: {
+    marker: '', // No explicit marker - uses possessive pronouns
+    markerPosition: 'after-object',
+    usePossessiveAdjectives: true,
+  },
+  roleMarkers: {
+    destination: {
+      primary: 'على',
+      alternatives: ['في', 'إلى', 'ب', 'قبل', 'بعد'],
+      position: 'before',
+    },
+    source: { primary: 'من', position: 'before' },
+    patient: { primary: '', position: 'before' },
+    style: { primary: 'بـ', alternatives: ['باستخدام'], position: 'before' },
+  },
+  tokenization: {
+    prefixes: ['ال', 'و', 'ف', 'ب', 'ك', 'ل'],
+  },
+} satisfies GrammarProfile;

package/src/languages/bn.ts

@@ -0,0 +1,45 @@
+/**
+ * Bengali Grammar
+ *
+ * SOV word order, postpositions (কে, তে, থেকে, etc.), Bengali script.
+ * Agglutinative language similar to Hindi.
+ */
+
+import type { GrammarProfile } from './types';
+
+export const bengaliGrammar = {
+  code: 'bn',
+  name: 'Bengali',
+  nativeName: 'বাংলা',
+  direction: 'ltr',
+  script: 'bengali',
+  wordOrder: 'SOV',
+  markingStrategy: 'postposition',
+  usesSpaces: true,
+  defaultVerbForm: 'imperative',
+  verb: {
+    position: 'end',
+    suffixes: ['ুন', 'ো', 'া', 'ে', 'ি'],
+    subjectDrop: true,
+  },
+  possessive: {
+    marker: 'র',
+    markerPosition: 'between',
+  },
+  roleMarkers: {
+    patient: { primary: 'কে', position: 'after' },
+    destination: { primary: 'তে', alternatives: ['এ'], position: 'after' },
+    source: { primary: 'থেকে', position: 'after' },
+    style: { primary: 'দিয়ে', position: 'after' },
+    event: { primary: 'তে', position: 'after' },
+  },
+  tokenization: {
+    particles: ['কে', 'তে', 'থেকে', 'র', 'এর', 'দিয়ে', 'জন্য', 'পর্যন্ত'],
+    boundaryStrategy: 'space',
+  },
+  // Bengali writes a case ending onto its noun (`#outকে`, `#outএ`; M2 N5).
+  // After a Bengali word only `কে`, which never changes: the locative is `তে`
+  // after a vowel and the vowel sign `ে` after a consonant (`ক্লিকে`), which
+  // the reader does not split.
+  attachedMarkers: { markers: ['কে', 'তে', 'এ'], afterWords: ['কে'] },
+} satisfies GrammarProfile;

package/src/languages/de.ts

@@ -0,0 +1,37 @@
+/**
+ * German Grammar
+ *
+ * SVO word order (V2 in main clauses), prepositions, space-separated.
+ * Features case system, compound words, and verb-second word order in main clauses.
+ */
+
+import type { GrammarProfile } from './types';
+
+export const germanGrammar = {
+  code: 'de',
+  name: 'German',
+  nativeName: 'Deutsch',
+  direction: 'ltr',
+  script: 'latin',
+  wordOrder: 'SVO',
+  markingStrategy: 'preposition',
+  usesSpaces: true,
+  // Infinitive is standard for German software UI (Speichern, Öffnen, Schließen)
+  // Also used in written instructions ("Bitte nicht stören")
+  defaultVerbForm: 'infinitive',
+  verb: {
+    position: 'start',
+    subjectDrop: false,
+  },
+  possessive: {
+    marker: '', // German uses possessive pronouns directly
+    markerPosition: 'before-property',
+    usePossessiveAdjectives: true,
+  },
+  roleMarkers: {
+    destination: { primary: 'auf', alternatives: ['zu', 'in'], position: 'before' },
+    source: { primary: 'von', alternatives: ['aus'], position: 'before' },
+    patient: { primary: '', position: 'before' },
+    style: { primary: 'mit', position: 'before' },
+  },
+} satisfies GrammarProfile;

package/src/languages/en.ts

@@ -0,0 +1,36 @@
+/**
+ * English Grammar
+ *
+ * SVO word order, prepositions, space-separated.
+ */
+
+import type { GrammarProfile } from './types';
+
+export const englishGrammar = {
+  code: 'en',
+  name: 'English',
+  nativeName: 'English',
+  direction: 'ltr',
+  script: 'latin',
+  wordOrder: 'SVO',
+  markingStrategy: 'preposition',
+  usesSpaces: true,
+  // English infinitive and imperative are identical ("toggle", "save", "open")
+  defaultVerbForm: 'base',
+  verb: {
+    position: 'start',
+    subjectDrop: false,
+  },
+  possessive: {
+    marker: "'s",
+    markerPosition: 'after-object',
+  },
+  roleMarkers: {
+    destination: { primary: 'on', alternatives: ['to', 'from'], position: 'before' },
+    source: { primary: 'from', position: 'before' },
+    patient: { primary: '', position: 'before' },
+    style: { primary: 'with', alternatives: ['by', 'using'], position: 'before' },
+    responseType: { primary: 'as', position: 'before' },
+    method: { primary: 'via', alternatives: ['using'], position: 'before' },
+  },
+} satisfies GrammarProfile;

package/src/languages/es.ts

@@ -0,0 +1,40 @@
+/**
+ * Spanish Grammar
+ *
+ * SVO word order, prepositions, space-separated.
+ * Features rich verb conjugation with pro-drop (subject omission).
+ */
+
+import type { GrammarProfile } from './types';
+
+export const spanishGrammar = {
+  code: 'es',
+  name: 'Spanish',
+  nativeName: 'Español',
+  direction: 'ltr',
+  script: 'latin',
+  wordOrder: 'SVO',
+  markingStrategy: 'preposition',
+  usesSpaces: true,
+  // Infinitive is standard for Spanish software UI (Guardar, Cancelar, Abrir)
+  // This matches macOS, Windows, and web app conventions
+  defaultVerbForm: 'infinitive',
+  verb: {
+    position: 'start',
+    subjectDrop: true,
+  },
+  possessive: {
+    marker: 'de', // Spanish uses "de" for general possession
+    markerPosition: 'before-property',
+    usePossessiveAdjectives: true,
+  },
+  roleMarkers: {
+    // `hacia` is the i18n grammar's optional destination render form ("towards");
+    // without it here a rendered/user `hacia` clause silently dropped the
+    // destination (add → default `me`, put → null parse). Vocab Batch 1 (V2+V4).
+    destination: { primary: 'en', alternatives: ['sobre', 'a', 'hacia'], position: 'before' },
+    source: { primary: 'de', alternatives: ['desde'], position: 'before' },
+    patient: { primary: '', position: 'before' },
+    style: { primary: 'con', position: 'before' },
+  },
+} satisfies GrammarProfile;

package/src/languages/fr.ts

@@ -0,0 +1,36 @@
+/**
+ * French Grammar
+ *
+ * SVO word order, prepositions, space-separated.
+ * Features rich verb conjugation and gendered articles.
+ */
+
+import type { GrammarProfile } from './types';
+
+export const frenchGrammar = {
+  code: 'fr',
+  name: 'French',
+  nativeName: 'Français',
+  direction: 'ltr',
+  script: 'latin',
+  wordOrder: 'SVO',
+  markingStrategy: 'preposition',
+  usesSpaces: true,
+  // Infinitive is standard for French software UI (Enregistrer, Ouvrir, Fermer)
+  defaultVerbForm: 'infinitive',
+  verb: {
+    position: 'start',
+    subjectDrop: false,
+  },
+  possessive: {
+    marker: 'de', // Uses "de" for general possession
+    markerPosition: 'before-property',
+    usePossessiveAdjectives: true,
+  },
+  roleMarkers: {
+    destination: { primary: 'sur', alternatives: ['à', 'dans'], position: 'before' },
+    source: { primary: 'de', alternatives: ['depuis'], position: 'before' },
+    patient: { primary: '', position: 'before' },
+    style: { primary: 'avec', position: 'before' },
+  },
+} satisfies GrammarProfile;

package/src/languages/he.ts

@@ -0,0 +1,38 @@
+/**
+ * Hebrew Grammar
+ *
+ * SVO word order, prepositions, RTL (right-to-left), space-separated.
+ * Hebrew uses prefix prepositions (ב, ל, מ, כ) that attach to the following word.
+ * Features optional vowel points (nikkud) which are typically omitted in modern text.
+ */
+
+import type { GrammarProfile } from './types';
+
+export const hebrewGrammar = {
+  code: 'he',
+  name: 'Hebrew',
+  nativeName: 'עברית',
+  direction: 'rtl',
+  script: 'hebrew',
+  wordOrder: 'SVO',
+  markingStrategy: 'preposition',
+  usesSpaces: true,
+  verb: {
+    position: 'second', // SVO - verb comes after subject
+    subjectDrop: true,
+  },
+  possessive: {
+    marker: 'של', // "of" - used for general possession
+    markerPosition: 'before-property',
+    usePossessiveAdjectives: true,
+  },
+  roleMarkers: {
+    destination: { primary: 'על', alternatives: ['ב', 'אל', 'ל'], position: 'before' },
+    source: { primary: 'מ', alternatives: ['מן', 'מאת'], position: 'before' },
+    patient: { primary: 'את', position: 'before' }, // Direct object marker
+    style: { primary: 'עם', alternatives: ['ב'], position: 'before' },
+  },
+  tokenization: {
+    prefixes: ['ה', 'ו', 'ב', 'כ', 'ל', 'מ', 'ש'], // Common Hebrew prefixes
+  },
+} satisfies GrammarProfile;

package/src/languages/hi.ts

@@ -0,0 +1,44 @@
+/**
+ * Hindi Grammar
+ *
+ * SOV word order, postpositions (को, में, पर, से, etc.), Devanagari script.
+ * Fusional language with verb conjugation patterns.
+ */
+
+import type { GrammarProfile } from './types';
+
+export const hindiGrammar = {
+  code: 'hi',
+  name: 'Hindi',
+  nativeName: 'हिन्दी',
+  direction: 'ltr',
+  script: 'devanagari',
+  wordOrder: 'SOV',
+  markingStrategy: 'postposition',
+  usesSpaces: true,
+  // Hindi uses imperative forms for commands in software UI
+  defaultVerbForm: 'imperative',
+  verb: {
+    position: 'end',
+    suffixes: ['ें', 'ा', 'ी', 'े', 'ो'],
+    subjectDrop: true,
+  },
+  possessive: {
+    marker: 'का',
+    markerPosition: 'between',
+  },
+  roleMarkers: {
+    patient: { primary: 'को', position: 'after' },
+    // `को` is the accusative/patient marker only — NOT a destination alternative.
+    // Listing it here made a `को`-marked patient mis-parse as a destination, the
+    // dominant source of phantom `into` commands in hi (precision audit 2026-06-15).
+    destination: { primary: 'में', alternatives: ['पर'], position: 'after' },
+    source: { primary: 'से', position: 'after' },
+    style: { primary: 'से', position: 'after' },
+    event: { primary: 'पर', position: 'after' },
+  },
+  tokenization: {
+    particles: ['को', 'में', 'पर', 'से', 'का', 'की', 'के', 'तक', 'ने'],
+    boundaryStrategy: 'space',
+  },
+} satisfies GrammarProfile;

package/src/languages/id.ts

@@ -0,0 +1,34 @@
+/**
+ * Indonesian Grammar
+ *
+ * SVO word order, prepositions, space-separated, agglutinative.
+ * Features affixation for verb derivation (me-, ber-, di-, -kan, -i).
+ */
+
+import type { GrammarProfile } from './types';
+
+export const indonesianGrammar = {
+  code: 'id',
+  name: 'Indonesian',
+  nativeName: 'Bahasa Indonesia',
+  direction: 'ltr',
+  script: 'latin',
+  wordOrder: 'SVO',
+  markingStrategy: 'preposition',
+  usesSpaces: true,
+  verb: {
+    position: 'start',
+    subjectDrop: true,
+  },
+  possessive: {
+    marker: '', // Indonesian: "X saya" (X of mine), possessor follows noun
+    markerPosition: 'after-object',
+    usePossessiveAdjectives: true,
+  },
+  roleMarkers: {
+    destination: { primary: 'pada', alternatives: ['ke', 'di'], position: 'before' },
+    source: { primary: 'dari', position: 'before' },
+    patient: { primary: '', position: 'before' },
+    style: { primary: 'dengan', position: 'before' },
+  },
+} satisfies GrammarProfile;

package/src/languages/index.ts

@@ -0,0 +1,42 @@
+/**
+ * Every language's grammar, one module each (`@lokascript/framework/languages/<code>`).
+ * Import the module you need: this barrel is for tooling and tests.
+ */
+
+export type {
+  WordOrder,
+  MarkingStrategy,
+  ScriptType,
+  RoleMarker,
+  VerbConfig,
+  VerbForm,
+  TokenizationConfig,
+  PossessiveGrammar,
+  AttachedMarkers,
+  GrammarProfile,
+} from './types';
+
+export { arabicGrammar } from './ar';
+export { bengaliGrammar } from './bn';
+export { germanGrammar } from './de';
+export { englishGrammar } from './en';
+export { spanishGrammar } from './es';
+export { frenchGrammar } from './fr';
+export { hebrewGrammar } from './he';
+export { hindiGrammar } from './hi';
+export { indonesianGrammar } from './id';
+export { italianGrammar } from './it';
+export { japaneseGrammar } from './ja';
+export { koreanGrammar } from './ko';
+export { malayGrammar } from './ms';
+export { polishGrammar } from './pl';
+export { portugueseGrammar } from './pt';
+export { quechuaGrammar } from './qu';
+export { russianGrammar } from './ru';
+export { swahiliGrammar } from './sw';
+export { thaiGrammar } from './th';
+export { tagalogGrammar } from './tl';
+export { turkishGrammar } from './tr';
+export { ukrainianGrammar } from './uk';
+export { vietnameseGrammar } from './vi';
+export { chineseGrammar } from './zh';

package/src/languages/it.ts

@@ -0,0 +1,38 @@
+/**
+ * Italian Grammar
+ *
+ * SVO word order, prepositions, space-separated.
+ * Features rich verb conjugation with pro-drop (subject omission).
+ * Very similar to Spanish but with Italian-specific vocabulary.
+ */
+
+import type { GrammarProfile } from './types';
+
+export const italianGrammar = {
+  code: 'it',
+  name: 'Italian',
+  nativeName: 'Italiano',
+  direction: 'ltr',
+  script: 'latin',
+  wordOrder: 'SVO',
+  markingStrategy: 'preposition',
+  usesSpaces: true,
+  // Infinitive is standard for Italian software UI (Salvare, Annullare, Aprire)
+  // This matches macOS, Windows, and web app conventions
+  defaultVerbForm: 'infinitive',
+  verb: {
+    position: 'start',
+    subjectDrop: true,
+  },
+  possessive: {
+    marker: 'di', // Italian uses "di" for general possession
+    markerPosition: 'before-property',
+    usePossessiveAdjectives: true,
+  },
+  roleMarkers: {
+    destination: { primary: 'in', alternatives: ['su', 'a'], position: 'before' },
+    source: { primary: 'da', alternatives: ['di'], position: 'before' },
+    patient: { primary: '', position: 'before' },
+    style: { primary: 'con', position: 'before' },
+  },
+} satisfies GrammarProfile;

package/src/languages/ja.ts

@@ -0,0 +1,49 @@
+/**
+ * Japanese Grammar
+ *
+ * SOV word order, particles (を, に, で, etc.), no spaces between words.
+ * Agglutinative language with rich verb conjugation.
+ */
+
+import type { GrammarProfile } from './types';
+
+export const japaneseGrammar = {
+  code: 'ja',
+  name: 'Japanese',
+  nativeName: '日本語',
+  direction: 'ltr',
+  script: 'cjk',
+  wordOrder: 'SOV',
+  markingStrategy: 'particle',
+  usesSpaces: false,
+  // Japanese uses verb stem/masu-stem form, no clear infinitive/imperative distinction
+  // for UI commands. Uses katakana loanwords (トグル) or native stems (切り替え)
+  defaultVerbForm: 'base',
+  verb: {
+    position: 'end',
+    suffixes: ['る', 'て', 'た', 'ます', 'ない'],
+    subjectDrop: true,
+  },
+  possessive: {
+    marker: 'の',
+    markerPosition: 'between',
+  },
+  roleMarkers: {
+    patient: { primary: 'を', position: 'after' },
+    destination: { primary: 'に', alternatives: ['へ', 'で', 'の'], position: 'after' },
+    source: { primary: 'から', position: 'after' },
+    style: { primary: 'で', position: 'after' },
+    event: { primary: 'を', position: 'after' }, // Event as object marker
+    // Possession marker for "X's Y" patterns
+    // Note: の is used between target and patient: #button の .active
+  },
+  tokenization: {
+    particles: ['を', 'に', 'で', 'から', 'の', 'が', 'は', 'も', 'へ', 'と'],
+    boundaryStrategy: 'particle',
+  },
+  // Japanese writes a particle onto its noun (`#outに`, `自分から`; M2 N5).
+  attachedMarkers: {
+    markers: ['を', 'に', 'へ', 'で', 'の', 'から', 'として', 'の間で'],
+    afterWords: 'all',
+  },
+} satisfies GrammarProfile;