@lokascript/semantic 1.2.1 → 2.0.0

package/dist/tokenizers/swahili.d.ts

@@ -0,0 +1,24 @@
+/**
+ * Swahili Tokenizer
+ *
+ * Tokenizes Swahili (Kiswahili) hyperscript input.
+ * Swahili characteristics:
+ * - SVO word order
+ * - Agglutinative morphology
+ * - Noun class prefixes (m-, wa-, ki-, vi-, etc.)
+ * - Verb prefixes for subject/object agreement
+ * - No grammatical gender, but noun classes
+ */
+import type { TokenKind, TokenStream } from '../types';
+import { BaseTokenizer } from './base';
+export declare class SwahiliTokenizer extends BaseTokenizer {
+    readonly language = "sw";
+    readonly direction: "ltr";
+    constructor();
+    tokenize(input: string): TokenStream;
+    classifyToken(token: string): TokenKind;
+    private extractWord;
+    private extractNumber;
+}
+export declare const swahiliTokenizer: SwahiliTokenizer;
+//# sourceMappingURL=swahili.d.ts.map

package/dist/tokenizers/swahili.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"swahili.d.ts","sourceRoot":"","sources":["../../src/tokenizers/swahili.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;GAUG;AAEH,OAAO,KAAK,EAAiB,SAAS,EAAE,WAAW,EAAE,MAAM,UAAU,CAAC;AACtE,OAAO,EACL,aAAa,EAWd,MAAM,QAAQ,CAAC;AAqGhB,qBAAa,gBAAiB,SAAQ,aAAa;IACjD,QAAQ,CAAC,QAAQ,QAAQ;IACzB,QAAQ,CAAC,SAAS,EAAG,KAAK,CAAU;;IAOpC,QAAQ,CAAC,KAAK,EAAE,MAAM,GAAG,WAAW;IA4FpC,aAAa,CAAC,KAAK,EAAE,MAAM,GAAG,SAAS;IAkBvC,OAAO,CAAC,WAAW;IAyBnB,OAAO,CAAC,aAAa;CA2CtB;AAED,eAAO,MAAM,gBAAgB,kBAAyB,CAAC"}

package/dist/tokenizers/thai.d.ts

@@ -0,0 +1,24 @@
+/**
+ * Thai Tokenizer
+ *
+ * Tokenizes Thai hyperscript input.
+ * Thai is an isolating SVO language with:
+ * - Thai script (U+0E00-U+0E7F)
+ * - No spaces between words (like Chinese/Japanese)
+ * - Prepositions for grammatical marking
+ * - CSS selectors are embedded ASCII
+ *
+ * This tokenizer derives keywords from the Thai profile (single source of truth)
+ * with extras for literals, positional words, and event names.
+ */
+import type { TokenKind, TokenStream } from '../types';
+import { BaseTokenizer } from './base';
+export declare class ThaiTokenizer extends BaseTokenizer {
+    readonly language = "th";
+    readonly direction: "ltr";
+    constructor();
+    tokenize(input: string): TokenStream;
+    classifyToken(value: string): TokenKind;
+}
+export declare const thaiTokenizer: ThaiTokenizer;
+//# sourceMappingURL=thai.d.ts.map

package/dist/tokenizers/thai.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"thai.d.ts","sourceRoot":"","sources":["../../src/tokenizers/thai.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;GAYG;AAEH,OAAO,KAAK,EAAiB,SAAS,EAAE,WAAW,EAAE,MAAM,UAAU,CAAC;AAEtE,OAAO,EACL,aAAa,EAWd,MAAM,QAAQ,CAAC;AAuDhB,qBAAa,aAAc,SAAQ,aAAa;IAC9C,QAAQ,CAAC,QAAQ,QAAQ;IACzB,QAAQ,CAAC,SAAS,EAAG,KAAK,CAAU;;IAQpC,QAAQ,CAAC,KAAK,EAAE,MAAM,GAAG,WAAW;IAyIpC,aAAa,CAAC,KAAK,EAAE,MAAM,GAAG,SAAS;CAgBxC;AAED,eAAO,MAAM,aAAa,eAAsB,CAAC"}

package/dist/tokenizers/tl.d.ts

@@ -0,0 +1,22 @@
+/**
+ * Tagalog Tokenizer
+ *
+ * Tokenizes Tagalog hyperscript input.
+ * Word order: VSO
+ * Direction: ltr
+ * Uses spaces: true
+ *
+ * This tokenizer derives keywords from the Tagalog profile (single source of truth)
+ * with extras for literals, positional words, and event names.
+ */
+import type { TokenKind, TokenStream } from '../types';
+import { BaseTokenizer } from './base';
+export declare class TagalogTokenizer extends BaseTokenizer {
+    readonly language = "tl";
+    readonly direction: "ltr";
+    constructor();
+    tokenize(input: string): TokenStream;
+    classifyToken(token: string): TokenKind;
+}
+export declare const tagalogTokenizer: TagalogTokenizer;
+//# sourceMappingURL=tl.d.ts.map

package/dist/tokenizers/tl.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"tl.d.ts","sourceRoot":"","sources":["../../src/tokenizers/tl.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;GAUG;AAEH,OAAO,KAAK,EAAiB,SAAS,EAAE,WAAW,EAAE,MAAM,UAAU,CAAC;AAEtE,OAAO,EACL,aAAa,EAUd,MAAM,QAAQ,CAAC;AAyChB,qBAAa,gBAAiB,SAAQ,aAAa;IACjD,QAAQ,CAAC,QAAQ,QAAQ;IACzB,QAAQ,CAAC,SAAS,EAAG,KAAK,CAAU;;IAQpC,QAAQ,CAAC,KAAK,EAAE,MAAM,GAAG,WAAW;IAoHpC,aAAa,CAAC,KAAK,EAAE,MAAM,GAAG,SAAS;CAgBxC;AAED,eAAO,MAAM,gBAAgB,kBAAyB,CAAC"}

package/dist/tokenizers/token-utils.d.ts

@@ -0,0 +1,97 @@
+/**
+ * Token Utilities
+ *
+ * Core token creation, stream implementation, and character classification.
+ * These are the foundational building blocks used by all tokenizers.
+ */
+import type { LanguageToken, TokenKind, TokenStream, StreamMark, SourcePosition } from '../types';
+/**
+ * Configuration for a native language time unit pattern.
+ * Used by tryNumberWithTimeUnits() to match language-specific time units.
+ */
+export interface TimeUnitMapping {
+    /** The pattern to match (e.g., 'segundos', 'ミリ秒') */
+    readonly pattern: string;
+    /** The standard suffix to use (ms, s, m, h) */
+    readonly suffix: string;
+    /** Length of the pattern (for optimization) */
+    readonly length: number;
+    /** Whether to check for word boundary after the pattern */
+    readonly checkBoundary?: boolean;
+    /** Character that cannot follow the pattern (e.g., 's' for 'm' to avoid 'ms') */
+    readonly notFollowedBy?: string;
+    /** Whether to do case-insensitive matching */
+    readonly caseInsensitive?: boolean;
+}
+/**
+ * Concrete implementation of TokenStream.
+ */
+export declare class TokenStreamImpl implements TokenStream {
+    readonly tokens: readonly LanguageToken[];
+    readonly language: string;
+    private pos;
+    constructor(tokens: LanguageToken[], language: string);
+    peek(offset?: number): LanguageToken | null;
+    advance(): LanguageToken;
+    isAtEnd(): boolean;
+    mark(): StreamMark;
+    reset(mark: StreamMark): void;
+    position(): number;
+    /**
+     * Get remaining tokens as an array.
+     */
+    remaining(): LanguageToken[];
+    /**
+     * Consume tokens while predicate is true.
+     */
+    takeWhile(predicate: (token: LanguageToken) => boolean): LanguageToken[];
+    /**
+     * Skip tokens while predicate is true.
+     */
+    skipWhile(predicate: (token: LanguageToken) => boolean): void;
+}
+/**
+ * Create a source position from start and end offsets.
+ */
+export declare function createPosition(start: number, end: number): SourcePosition;
+/**
+ * Options for creating a token with optional morphological data.
+ */
+export interface CreateTokenOptions {
+    /** Explicitly normalized form from keyword map */
+    normalized?: string;
+    /** Morphologically normalized stem */
+    stem?: string;
+    /** Confidence in the stem (0.0-1.0) */
+    stemConfidence?: number;
+}
+/**
+ * Create a language token.
+ */
+export declare function createToken(value: string, kind: TokenKind, position: SourcePosition, normalizedOrOptions?: string | CreateTokenOptions): LanguageToken;
+/**
+ * Check if a character is whitespace.
+ */
+export declare function isWhitespace(char: string): boolean;
+/**
+ * Check if a string starts with a CSS selector prefix.
+ * Includes JSX-style element selectors: <form />, <div>
+ */
+export declare function isSelectorStart(char: string): boolean;
+/**
+ * Check if a character is a quote (string delimiter).
+ */
+export declare function isQuote(char: string): boolean;
+/**
+ * Check if a character is a digit.
+ */
+export declare function isDigit(char: string): boolean;
+/**
+ * Check if a character is an ASCII letter.
+ */
+export declare function isAsciiLetter(char: string): boolean;
+/**
+ * Check if a character is part of an ASCII identifier.
+ */
+export declare function isAsciiIdentifierChar(char: string): boolean;
+//# sourceMappingURL=token-utils.d.ts.map

package/dist/tokenizers/token-utils.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"token-utils.d.ts","sourceRoot":"","sources":["../../src/tokenizers/token-utils.ts"],"names":[],"mappings":"AAAA;;;;;GAKG;AAEH,OAAO,KAAK,EAAE,aAAa,EAAE,SAAS,EAAE,WAAW,EAAE,UAAU,EAAE,cAAc,EAAE,MAAM,UAAU,CAAC;AAMlG;;;GAGG;AACH,MAAM,WAAW,eAAe;IAC9B,qDAAqD;IACrD,QAAQ,CAAC,OAAO,EAAE,MAAM,CAAC;IACzB,+CAA+C;IAC/C,QAAQ,CAAC,MAAM,EAAE,MAAM,CAAC;IACxB,+CAA+C;IAC/C,QAAQ,CAAC,MAAM,EAAE,MAAM,CAAC;IACxB,2DAA2D;IAC3D,QAAQ,CAAC,aAAa,CAAC,EAAE,OAAO,CAAC;IACjC,iFAAiF;IACjF,QAAQ,CAAC,aAAa,CAAC,EAAE,MAAM,CAAC;IAChC,8CAA8C;IAC9C,QAAQ,CAAC,eAAe,CAAC,EAAE,OAAO,CAAC;CACpC;AAMD;;GAEG;AACH,qBAAa,eAAgB,YAAW,WAAW;IACjD,QAAQ,CAAC,MAAM,EAAE,SAAS,aAAa,EAAE,CAAC;IAC1C,QAAQ,CAAC,QAAQ,EAAE,MAAM,CAAC;IAC1B,OAAO,CAAC,GAAG,CAAa;gBAEZ,MAAM,EAAE,aAAa,EAAE,EAAE,QAAQ,EAAE,MAAM;IAKrD,IAAI,CAAC,MAAM,GAAE,MAAU,GAAG,aAAa,GAAG,IAAI;IAQ9C,OAAO,IAAI,aAAa;IAOxB,OAAO,IAAI,OAAO;IAIlB,IAAI,IAAI,UAAU;IAIlB,KAAK,CAAC,IAAI,EAAE,UAAU,GAAG,IAAI;IAI7B,QAAQ,IAAI,MAAM;IAIlB;;OAEG;IACH,SAAS,IAAI,aAAa,EAAE;IAI5B;;OAEG;IACH,SAAS,CAAC,SAAS,EAAE,CAAC,KAAK,EAAE,aAAa,KAAK,OAAO,GAAG,aAAa,EAAE;IAQxE;;OAEG;IACH,SAAS,CAAC,SAAS,EAAE,CAAC,KAAK,EAAE,aAAa,KAAK,OAAO,GAAG,IAAI;CAK9D;AAMD;;GAEG;AACH,wBAAgB,cAAc,CAAC,KAAK,EAAE,MAAM,EAAE,GAAG,EAAE,MAAM,GAAG,cAAc,CAEzE;AAED;;GAEG;AACH,MAAM,WAAW,kBAAkB;IACjC,kDAAkD;IAClD,UAAU,CAAC,EAAE,MAAM,CAAC;IACpB,sCAAsC;IACtC,IAAI,CAAC,EAAE,MAAM,CAAC;IACd,uCAAuC;IACvC,cAAc,CAAC,EAAE,MAAM,CAAC;CACzB;AAED;;GAEG;AACH,wBAAgB,WAAW,CACzB,KAAK,EAAE,MAAM,EACb,IAAI,EAAE,SAAS,EACf,QAAQ,EAAE,cAAc,EACxB,mBAAmB,CAAC,EAAE,MAAM,GAAG,kBAAkB,GAChD,aAAa,CAoBf;AAED;;GAEG;AACH,wBAAgB,YAAY,CAAC,IAAI,EAAE,MAAM,GAAG,OAAO,CAElD;AAED;;;GAGG;AACH,wBAAgB,eAAe,CAAC,IAAI,EAAE,MAAM,GAAG,OAAO,CAIrD;AAED;;GAEG;AACH,wBAAgB,OAAO,CAAC,IAAI,EAAE,MAAM,GAAG,OAAO,CAE7C;AAED;;GAEG;AACH,wBAAgB,OAAO,CAAC,IAAI,EAAE,MAAM,GAAG,OAAO,CAE7C;AAED;;GAEG;AACH,wBAAgB,aAAa,CAAC,IAAI,EAAE,MAAM,GAAG,OAAO,CAEnD;AAED;;GAEG;AACH,wBAAgB,qBAAqB,CAAC,IAAI,EAAE,MAAM,GAAG,OAAO,CAE3D"}

package/dist/tokenizers/turkish.d.ts

@@ -0,0 +1,42 @@
+/**
+ * Turkish Tokenizer
+ *
+ * Tokenizes Turkish hyperscript input.
+ * Turkish is challenging because:
+ * - Highly agglutinative (many suffixes attach to words)
+ * - Strict vowel harmony rules
+ * - Postpositions instead of prepositions
+ * - No grammatical gender
+ * - Word order is typically SOV
+ */
+import type { TokenKind, TokenStream } from '../types';
+import { BaseTokenizer } from './base';
+export declare class TurkishTokenizer extends BaseTokenizer {
+    readonly language = "tr";
+    readonly direction: "ltr";
+    constructor();
+    tokenize(input: string): TokenStream;
+    classifyToken(token: string): TokenKind;
+    /**
+     * Try to match multi-word phrases that function as single units.
+     * Multi-word phrases are included in profileKeywords and sorted longest-first,
+     * so they'll be matched before their constituent words.
+     *
+     * Examples: "üzerine gelme" (hover), "fare üzerinde" (mouseover)
+     */
+    private tryMultiWordPhrase;
+    /**
+     * Extract a Turkish word.
+     * Uses morphological normalization to handle verb conjugations.
+     */
+    private extractTurkishWord;
+    /**
+     * Extract a number, including Turkish time unit suffixes.
+     */
+    private extractTurkishNumber;
+}
+/**
+ * Singleton instance.
+ */
+export declare const turkishTokenizer: TurkishTokenizer;
+//# sourceMappingURL=turkish.d.ts.map

package/dist/tokenizers/turkish.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"turkish.d.ts","sourceRoot":"","sources":["../../src/tokenizers/turkish.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;GAUG;AAEH,OAAO,KAAK,EAAiB,SAAS,EAAE,WAAW,EAAE,MAAM,UAAU,CAAC;AACtE,OAAO,EACL,aAAa,EAYd,MAAM,QAAQ,CAAC;AAiLhB,qBAAa,gBAAiB,SAAQ,aAAa;IACjD,QAAQ,CAAC,QAAQ,QAAQ;IACzB,QAAQ,CAAC,SAAS,EAAG,KAAK,CAAU;;IASpC,QAAQ,CAAC,KAAK,EAAE,MAAM,GAAG,WAAW;IAkGpC,aAAa,CAAC,KAAK,EAAE,MAAM,GAAG,SAAS;IAoBvC;;;;;;OAMG;IACH,OAAO,CAAC,kBAAkB;IA6B1B;;;OAGG;IACH,OAAO,CAAC,kBAAkB;IAsC1B;;OAEG;IACH,OAAO,CAAC,oBAAoB;CAM7B;AAED;;GAEG;AACH,eAAO,MAAM,gBAAgB,kBAAyB,CAAC"}

package/dist/tokenizers/ukrainian.d.ts

@@ -0,0 +1,25 @@
+/**
+ * Ukrainian Tokenizer
+ *
+ * Tokenizes Ukrainian hyperscript input.
+ * Ukrainian characteristics:
+ * - SVO word order (relatively free, but SVO is common)
+ * - Space-separated words
+ * - Prepositions
+ * - Cyrillic alphabet (with unique letters і, ї, є, ґ)
+ * - Fusional morphology with verb conjugations
+ * - Imperative form used in software UI (infinitive also common)
+ */
+import type { TokenKind, TokenStream } from '../types';
+import { BaseTokenizer } from './base';
+export declare class UkrainianTokenizer extends BaseTokenizer {
+    readonly language = "uk";
+    readonly direction: "ltr";
+    constructor();
+    tokenize(input: string): TokenStream;
+    classifyToken(token: string): TokenKind;
+    private extractWord;
+    private extractNumber;
+}
+export declare const ukrainianTokenizer: UkrainianTokenizer;
+//# sourceMappingURL=ukrainian.d.ts.map

package/dist/tokenizers/ukrainian.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"ukrainian.d.ts","sourceRoot":"","sources":["../../src/tokenizers/ukrainian.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;GAWG;AAEH,OAAO,KAAK,EAAiB,SAAS,EAAE,WAAW,EAAE,MAAM,UAAU,CAAC;AACtE,OAAO,EACL,aAAa,EAWd,MAAM,QAAQ,CAAC;AA+HhB,qBAAa,kBAAmB,SAAQ,aAAa;IACnD,QAAQ,CAAC,QAAQ,QAAQ;IACzB,QAAQ,CAAC,SAAS,EAAG,KAAK,CAAU;;IASpC,QAAQ,CAAC,KAAK,EAAE,MAAM,GAAG,WAAW;IA4FpC,aAAa,CAAC,KAAK,EAAE,MAAM,GAAG,SAAS;IAkBvC,OAAO,CAAC,WAAW;IA0BnB,OAAO,CAAC,aAAa;CAiFtB;AAED,eAAO,MAAM,kBAAkB,oBAA2B,CAAC"}

package/dist/tokenizers/vietnamese.d.ts

@@ -0,0 +1,47 @@
+/**
+ * Vietnamese Tokenizer
+ *
+ * Tokenizes Vietnamese hyperscript input.
+ * Vietnamese is an isolating (analytic) language with:
+ * - SVO word order (like English)
+ * - Latin script with extensive diacritics (tone marks)
+ * - No verb conjugation or noun declension
+ * - Space-separated syllables (can be multi-syllable words)
+ * - Prepositions for grammatical roles
+ *
+ * Vietnamese diacritics:
+ * - Tone marks: à á ả ã ạ (and similar for other vowels)
+ * - Vowel modifications: ă â ê ô ơ ư đ
+ *
+ * Examples:
+ *   chuyển đổi .active  → toggle .active
+ *   thêm .highlight     → add .highlight
+ *   hiển thị #modal     → show #modal
+ */
+import type { TokenKind, TokenStream } from '../types';
+import { BaseTokenizer } from './base';
+export declare class VietnameseTokenizer extends BaseTokenizer {
+    readonly language = "vi";
+    readonly direction: "ltr";
+    constructor();
+    tokenize(input: string): TokenStream;
+    classifyToken(token: string): TokenKind;
+    /**
+     * Try to match a multi-word phrase.
+     * Multi-word phrases are included in profileKeywords and sorted longest-first.
+     */
+    private tryMultiWordPhrase;
+    /**
+     * Extract a Vietnamese word (single syllable/word).
+     */
+    private extractVietnameseWord;
+    /**
+     * Extract a number, including time unit suffixes.
+     */
+    private extractVietnameseNumber;
+}
+/**
+ * Singleton instance.
+ */
+export declare const vietnameseTokenizer: VietnameseTokenizer;
+//# sourceMappingURL=vietnamese.d.ts.map

package/dist/tokenizers/vietnamese.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"vietnamese.d.ts","sourceRoot":"","sources":["../../src/tokenizers/vietnamese.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;GAmBG;AAEH,OAAO,KAAK,EAAiB,SAAS,EAAE,WAAW,EAAE,MAAM,UAAU,CAAC;AACtE,OAAO,EACL,aAAa,EAWd,MAAM,QAAQ,CAAC;AA0HhB,qBAAa,mBAAoB,SAAQ,aAAa;IACpD,QAAQ,CAAC,QAAQ,QAAQ;IACzB,QAAQ,CAAC,SAAS,EAAG,KAAK,CAAU;;IAQpC,QAAQ,CAAC,KAAK,EAAE,MAAM,GAAG,WAAW;IA0GpC,aAAa,CAAC,KAAK,EAAE,MAAM,GAAG,SAAS;IAmBvC;;;OAGG;IACH,OAAO,CAAC,kBAAkB;IAyB1B;;OAEG;IACH,OAAO,CAAC,qBAAqB;IA2B7B;;OAEG;IACH,OAAO,CAAC,uBAAuB;CA0DhC;AAED;;GAEG;AACH,eAAO,MAAM,mBAAmB,qBAA4B,CAAC"}

package/dist/types/grammar-types.d.ts

@@ -0,0 +1,224 @@
+/**
+ * Grammar Types for Semantic Multilingual Parsing
+ *
+ * These types define the semantic role system used across all 13 supported languages.
+ * Originally from @lokascript/i18n, now consolidated here for package independence.
+ *
+ * Key Linguistic Concepts:
+ * - Word Order: SVO, SOV, VSO (and variations)
+ * - Adposition Type: Preposition (English) vs Postposition (Japanese/Korean)
+ * - Morphology: Isolating (Chinese) vs Agglutinative (Turkish) vs Fusional (Arabic)
+ * - Text Direction: LTR vs RTL
+ */
+/**
+ * Semantic roles in hyperscript commands.
+ * These are universal across all 13 supported languages - only the surface form changes.
+ *
+ * ## Core Thematic Roles (from linguistic theory)
+ * | Role        | Usage | Purpose                     | Example                   |
+ * |-------------|-------|-----------------------------|---------------------------|
+ * | action      | 100%  | Command verb                | toggle, put, fetch        |
+ * | patient     | 90%   | What is acted upon          | .active, #count           |
+ * | destination | 40%   | Where something goes        | into #output, to .class   |
+ * | source      | 13%   | Where something comes from  | from #input, from URL     |
+ * | event       | 106%  | Trigger events              | click, keydown, submit    |
+ * | condition   | 8%    | Boolean expressions         | if x > 5, when visible    |
+ * | agent       | 0%    | Who performs action         | Reserved for future use   |
+ * | goal        | 1%    | Target value/state          | to 'red' (in transition)  |
+ *
+ * ## Quantitative Roles (answer "how much/long")
+ * | Role     | Usage | Purpose        | Example              |
+ * |----------|-------|----------------|----------------------|
+ * | quantity | 7%    | Numeric amount | by 5, 3 times        |
+ * | duration | 1%    | Time span      | for 5 seconds, 500ms |
+ *
+ * ## Adverbial/Modifier Roles (answer "how/by what means")
+ * | Role         | Usage | Purpose                   | Example           |
+ * |--------------|-------|---------------------------|-------------------|
+ * | style        | 2%    | Animation/behavior        | with fade         |
+ * | manner       | 2%    | Insertion position        | before, after     |
+ * | method       | 1%    | HTTP method/technique     | via POST, as GET  |
+ * | responseType | 1%    | Response format           | as json, as html  |
+ *
+ * ## Control Flow Roles
+ * | Role     | Usage | Purpose      | Example               |
+ * |----------|-------|--------------|-----------------------|
+ * | loopType | 6%    | Loop variant | forever, until, times |
+ *
+ * ## Design Notes
+ * - Low-usage roles (agent, goal, method, responseType) are intentionally kept for:
+ *   - Linguistic completeness across all 13 languages
+ *   - Future extensibility (AI agents, server-side execution)
+ *   - Command-specific semantics (fetch, transition)
+ * - Each role has distinct grammatical markers per language (see profiles/index.ts)
+ * - Usage percentages based on pattern database analysis
+ */
+export type SemanticRole = 'action' | 'agent' | 'patient' | 'source' | 'destination' | 'goal' | 'event' | 'condition' | 'quantity' | 'duration' | 'responseType' | 'method' | 'style' | 'manner' | 'loopType' | 'continues';
+/**
+ * Word order patterns
+ * These represent the major typological categories
+ */
+export type WordOrder = 'SVO' | 'SOV' | 'VSO' | 'VOS' | 'OVS' | 'OSV' | 'free';
+/**
+ * Where grammatical markers appear relative to their noun/verb
+ */
+export type AdpositionType = 'preposition' | 'postposition' | 'circumposition' | 'none';
+/**
+ * Morphological typology - how words are constructed
+ */
+export type MorphologyType = 'isolating' | 'agglutinative' | 'fusional' | 'polysynthetic';
+/**
+ * A grammatical marker (particle, case ending, preposition)
+ */
+export interface GrammaticalMarker {
+    form: string;
+    role: SemanticRole;
+    position: AdpositionType;
+    required: boolean;
+    alternatives?: string[];
+}
+/**
+ * Complete grammatical profile for a language
+ * This captures the essential typological features needed for transformation
+ */
+export interface LanguageProfile {
+    code: string;
+    name: string;
+    wordOrder: WordOrder;
+    adpositionType: AdpositionType;
+    morphology: MorphologyType;
+    direction: 'ltr' | 'rtl';
+    markers: GrammaticalMarker[];
+    canonicalOrder: SemanticRole[];
+    rules?: GrammarRule[];
+}
+/**
+ * Pattern for transforming hyperscript structures
+ */
+export interface GrammarRule {
+    name: string;
+    description: string;
+    match: PatternMatcher;
+    transform: PatternTransform;
+    priority: number;
+}
+/**
+ * Matches a hyperscript pattern
+ */
+export interface PatternMatcher {
+    commands?: string[];
+    requiredRoles: SemanticRole[];
+    optionalRoles?: SemanticRole[];
+    predicate?: (parsed: ParsedStatement) => boolean;
+}
+/**
+ * Defines how to transform a matched pattern
+ */
+export interface PatternTransform {
+    roleOrder: SemanticRole[];
+    insertMarkers?: boolean;
+    custom?: (parsed: ParsedStatement, profile: LanguageProfile) => string;
+}
+/**
+ * A parsed hyperscript statement broken into semantic components
+ */
+export interface ParsedStatement {
+    type: 'event-handler' | 'command' | 'conditional' | 'loop';
+    roles: Map<SemanticRole, ParsedElement>;
+    original: string;
+}
+/**
+ * A single element with its semantic role
+ */
+export interface ParsedElement {
+    role: SemanticRole;
+    value: string;
+    translated?: string;
+    isSelector?: boolean;
+    isLiteral?: boolean;
+}
+/**
+ * Universal templates for common hyperscript patterns
+ * These define the semantic structure independent of surface form
+ */
+export declare const UNIVERSAL_PATTERNS: {
+    readonly eventIncrement: {
+        readonly name: "event-increment";
+        readonly roles: SemanticRole[];
+        readonly english: "on {event} {action} {patient}";
+    };
+    readonly putInto: {
+        readonly name: "put-into";
+        readonly roles: SemanticRole[];
+        readonly english: "{action} {patient} into {destination}";
+    };
+    readonly addTo: {
+        readonly name: "add-to";
+        readonly roles: SemanticRole[];
+        readonly english: "{action} {patient} to {destination}";
+    };
+    readonly toggleOn: {
+        readonly name: "toggle-on";
+        readonly roles: SemanticRole[];
+        readonly english: "{action} {patient} on {destination}";
+    };
+    readonly waitDuration: {
+        readonly name: "wait-duration";
+        readonly roles: SemanticRole[];
+        readonly english: "{action} {quantity}";
+    };
+    readonly conditional: {
+        readonly name: "conditional";
+        readonly roles: SemanticRole[];
+        readonly english: "{action} {condition} then ... end";
+    };
+    readonly fetchAs: {
+        readonly name: "fetch-as";
+        readonly roles: SemanticRole[];
+        readonly english: "{action} {source} as {method}";
+    };
+    readonly showWith: {
+        readonly name: "show-with";
+        readonly roles: SemanticRole[];
+        readonly english: "{action} {patient} with {style}";
+    };
+    readonly transitionOver: {
+        readonly name: "transition-over";
+        readonly roles: SemanticRole[];
+        readonly english: "{action} {patient} over {duration}";
+    };
+};
+/**
+ * Default profiles for major language families
+ * Individual languages inherit and override these
+ */
+export declare const LANGUAGE_FAMILY_DEFAULTS: Record<string, Partial<LanguageProfile>>;
+/**
+ * Reorder semantic roles according to target language.
+ * Includes a safety net to append any roles present in input
+ * but missing from the target order, preventing data loss.
+ */
+export declare function reorderRoles(roles: Map<SemanticRole, ParsedElement>, targetOrder: SemanticRole[]): ParsedElement[];
+/**
+ * Insert grammatical markers between elements
+ */
+export declare function insertMarkers(elements: ParsedElement[], markers: GrammaticalMarker[], adpositionType: AdpositionType): string[];
+/**
+ * Intelligently joins tokens, handling agglutinative suffixes and prefixes.
+ *
+ * Rules:
+ * 1. If a token ends with '-' (prefix marker), no space after it
+ * 2. If a token starts with '-' (suffix marker), no space before it
+ * 3. Removes the hyphen indicators from the final output
+ *
+ * Examples:
+ * - ['#count', '-ta'] → '#countta' (Quechua accusative suffix)
+ * - ['بـ-', 'الماوس'] → 'بـالماوس' (Arabic prefix attachment)
+ * - ['value', 'を'] → 'value を' (Japanese particle, normal spacing)
+ */
+export declare function joinTokens(tokens: string[]): string;
+/**
+ * Transform a parsed statement to target language
+ */
+export declare function transformStatement(parsed: ParsedStatement, _sourceProfile: LanguageProfile, targetProfile: LanguageProfile): string;
+//# sourceMappingURL=grammar-types.d.ts.map

package/dist/types/grammar-types.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"grammar-types.d.ts","sourceRoot":"","sources":["../../src/types/grammar-types.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;GAWG;AAMH;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;GA0CG;AACH,MAAM,MAAM,YAAY,GAEpB,QAAQ,GACR,OAAO,GACP,SAAS,GACT,QAAQ,GACR,aAAa,GACb,MAAM,GACN,OAAO,GACP,WAAW,GAEX,UAAU,GACV,UAAU,GAEV,cAAc,GACd,QAAQ,GACR,OAAO,GACP,QAAQ,GAER,UAAU,GAEV,WAAW,CAAC;AAEhB;;;GAGG;AACH,MAAM,MAAM,SAAS,GAAG,KAAK,GAAG,KAAK,GAAG,KAAK,GAAG,KAAK,GAAG,KAAK,GAAG,KAAK,GAAG,MAAM,CAAC;AAE/E;;GAEG;AACH,MAAM,MAAM,cAAc,GAAG,aAAa,GAAG,cAAc,GAAG,gBAAgB,GAAG,MAAM,CAAC;AAExF;;GAEG;AACH,MAAM,MAAM,cAAc,GACtB,WAAW,GACX,eAAe,GACf,UAAU,GACV,eAAe,CAAC;AAEpB;;GAEG;AACH,MAAM,WAAW,iBAAiB;IAChC,IAAI,EAAE,MAAM,CAAC;IACb,IAAI,EAAE,YAAY,CAAC;IACnB,QAAQ,EAAE,cAAc,CAAC;IACzB,QAAQ,EAAE,OAAO,CAAC;IAClB,YAAY,CAAC,EAAE,MAAM,EAAE,CAAC;CACzB;AAMD;;;GAGG;AACH,MAAM,WAAW,eAAe;IAC9B,IAAI,EAAE,MAAM,CAAC;IACb,IAAI,EAAE,MAAM,CAAC;IAGb,SAAS,EAAE,SAAS,CAAC;IACrB,cAAc,EAAE,cAAc,CAAC;IAC/B,UAAU,EAAE,cAAc,CAAC;IAC3B,SAAS,EAAE,KAAK,GAAG,KAAK,CAAC;IAGzB,OAAO,EAAE,iBAAiB,EAAE,CAAC;IAK7B,cAAc,EAAE,YAAY,EAAE,CAAC;IAG/B,KAAK,CAAC,EAAE,WAAW,EAAE,CAAC;CACvB;AAED;;GAEG;AACH,MAAM,WAAW,WAAW;IAC1B,IAAI,EAAE,MAAM,CAAC;IACb,WAAW,EAAE,MAAM,CAAC;IAGpB,KAAK,EAAE,cAAc,CAAC;IAGtB,SAAS,EAAE,gBAAgB,CAAC;IAG5B,QAAQ,EAAE,MAAM,CAAC;CAClB;AAED;;GAEG;AACH,MAAM,WAAW,cAAc;IAE7B,QAAQ,CAAC,EAAE,MAAM,EAAE,CAAC;IAGpB,aAAa,EAAE,YAAY,EAAE,CAAC;IAG9B,aAAa,CAAC,EAAE,YAAY,EAAE,CAAC;IAG/B,SAAS,CAAC,EAAE,CAAC,MAAM,EAAE,eAAe,KAAK,OAAO,CAAC;CAClD;AAED;;GAEG;AACH,MAAM,WAAW,gBAAgB;IAE/B,SAAS,EAAE,YAAY,EAAE,CAAC;IAG1B,aAAa,CAAC,EAAE,OAAO,CAAC;IAGxB,MAAM,CAAC,EAAE,CAAC,MAAM,EAAE,eAAe,EAAE,OAAO,EAAE,eAAe,KAAK,MAAM,CAAC;CACxE;AAMD;;GAEG;AACH,MAAM,WAAW,eAAe;IAC9B,IAAI,EAAE,eAAe,GAAG,SAAS,GAAG,aAAa,GAAG,MAAM,CAAC;IAC3D,KAAK,EAAE,GAAG,CAAC,YAAY,EAAE,aAAa,CAAC,CAAC;IACxC,QAAQ,EAAE,MAAM,CAAC;CAClB;AAED;;GAEG;AACH,MAAM,WAAW,aAAa;IAC5B,IAAI,EAAE,YAAY,CAAC;IACnB,KAAK,EAAE,MAAM,CAAC;IACd,UAAU,CAAC,EAAE,MAAM,CAAC;IACpB,UAAU,CAAC,EAAE,OAAO,CAAC;IACrB,SAAS,CAAC,EAAE,OAAO,CAAC;CACrB;AAMD;;;GAGG;AACH,eAAO,MAAM,kBAAkB;;;wBAIc,YAAY,EAAE;;;;;wBAOR,YAAY,EAAE;;;;;wBAOd,YAAY,EAAE;;;;;wBAOd,YAAY,EAAE;;;;;wBAO5B,YAAY,EAAE;;;;;wBAOb,YAAY,EAAE;;;;;wBAOP,YAAY,EAAE;;;;;wBAOd,YAAY,EAAE;;;;;wBAOX,YAAY,EAAE;;;CAGpD,CAAC;AAMX;;;GAGG;AACH,eAAO,MAAM,wBAAwB,EAAE,MAAM,CAAC,MAAM,EAAE,OAAO,CAAC,eAAe,CAAC,CAqL7E,CAAC;AAMF;;;;GAIG;AACH,wBAAgB,YAAY,CAC1B,KAAK,EAAE,GAAG,CAAC,YAAY,EAAE,aAAa,CAAC,EACvC,WAAW,EAAE,YAAY,EAAE,GAC1B,aAAa,EAAE,CAsBjB;AAED;;GAEG;AACH,wBAAgB,aAAa,CAC3B,QAAQ,EAAE,aAAa,EAAE,EACzB,OAAO,EAAE,iBAAiB,EAAE,EAC5B,cAAc,EAAE,cAAc,GAC7B,MAAM,EAAE,CAwBV;AAED;;;;;;;;;;;;GAYG;AACH,wBAAgB,UAAU,CAAC,MAAM,EAAE,MAAM,EAAE,GAAG,MAAM,CAmCnD;AAED;;GAEG;AACH,wBAAgB,kBAAkB,CAChC,MAAM,EAAE,eAAe,EACvB,cAAc,EAAE,eAAe,EAC/B,aAAa,EAAE,eAAe,GAC7B,MAAM,CAUR"}