/**
 * Removes Indonesian affixes (prefixes and suffixes) from a word.
 *
 * Strips common Indonesian morphological affixes such as:
 * - Prefixes: me-, di-, ber-, pe-, pe-, ku-, kau-, men-, men-, per-, dll
 * - Suffixes: -kan, -an, -nya, -lah, -tah, -pun
 *
 * Used for text normalization, search indexing, and linguistic analysis.
 *
 * @param text - Word or text to strip affixes from
 * @returns Text with affixes removed
 *
 * @example
 * ```typescript
 * stemText('mempertanggungjawabkan'); // 'tanggung jawab'
 * stemText('berbicara');               // 'bicara'
 * stemText('mahasiswanya');             // 'mahasisw'
 * ```
 *
 * @example
 * For search optimization:
 * ```typescript
 * const query = 'mempertanggungjawabkan';
 * const normalized = stemText(query); // 'tanggung jawab'
 * // Use normalized for Indonesian full-text search
 * ```
 */
declare function stemText(text: string): string;

/**
 * Encodes a name into a phonetic code for fuzzy matching.
 *
 * Uses Soundex-like algorithm adapted for Indonesian names.
 * Handles Dutch and Arabic name normalizations for better
 * matching accuracy in Indonesian context.
 *
 * @param text - Name to encode (e.g., person name)
 * @returns Phonetic code (1 letter + 3 digits, e.g., 'S635')
 *
 * @example
 * ```typescript
 * encodePhonetic('Syahruddin'); // 'S635'
 * encodePhonetic('Ahmad');     // 'A530'
 * ```
 *
 * @example
 * For fuzzy name matching:
 * ```typescript
 * const code1 = encodePhonetic('Budi');
 * const code2 = encodePhonetic('Budi'); // same code = match
 * // Useful for deduping names with spelling variations
 * ```
 */
declare function encodePhonetic(text: string): string;
/**
 * Checks if two names match phonetically.
 *
 * Compares the phonetic codes of two names to determine
 * if they likely refer to the same person despite spelling
 * variations (typos, nicknames, alternative spellings).
 *
 * @param text1 - First name to compare
 * @param text2 - Second name to compare
 * @returns `true` if names match phonetically, `false` otherwise
 *
 * @example
 * ```typescript
 * isPhoneticMatch('Syahruddin', 'Syahrudin'); // true
 * isPhoneticMatch('Budi', 'Budi');           // true
 * isPhoneticMatch('Andi', 'Budi');            // false
 * ```
 *
 * @example
 * For search-as-you-type:
 * ```typescript
 * const typed = 'Syahrudin';
 * const candidates = ['Syahruddin', 'Budi', 'Andi'];
 * const match = candidates.find(c => isPhoneticMatch(typed, c));
 * // Returns 'Syahruddin'
 * ```
 */
declare function isPhoneticMatch(text1: string, text2: string): boolean;

/**
 * Splits Indonesian text into sentences.
 *
 * Handles Indonesian abbreviations (Yth., Bpk., Sdr., dll)
 * so periods inside abbreviations don't create false sentence breaks.
 *
 * @param text - Text to tokenize into sentences
 * @returns Array of sentences (empty array if input is empty/invalid)
 *
 * @example
 * ```typescript
 * tokenizeIndo("Kpd Yth. Bpk. Budi."); // ["Kpd Yth. Bpk. Budi."]
 * tokenizeIndo("Halo. Selamat pagi."); // ["Halo.", "Selamat pagi."]
 * ```
 *
 * @example
 * For processing formal Indonesian letters:
 * ```typescript
 * const letter = "Hormat kami, PT ABC.Jl. Merdeka No.10";
 * const sentences = tokenizeIndo(letter);
 * // Properly handles "Jl." (Jalan) abbreviation
 * ```
 */
declare function tokenizeIndo(text: string): string[];

/**
 * Collapses multiple spaces to single space and trims.
 *
 * @param text - Text to normalize
 * @returns Text with normalized whitespace
 *
 * @example
 * ```typescript
 * normalizeWhitespace("  Budi   pergi  ke   sekolah  ")
 * // "Budi pergi ke sekolah"
 * ```
 */
declare function normalizeWhitespace(text: string): string;
/**
 * Removes all non-letter/number characters except spaces.
 *
 * @param text - Text to clean
 * @returns Text with only letters, numbers, and spaces
 *
 * @example
 * ```typescript
 * stripNonAlphanumeric("Budi123@#$%"); // "Budi123"
 * stripNonAlphanumeric("Hello! World?"); // "Hello World"
 * ```
 */
declare function stripNonAlphanumeric(text: string): string;

/**
 * NLP Engine types for Indonesian text processing.
 *
 * @module nlp
 */
/**
 * Phonetic encoding result with metadata.
 */
interface PhoneticResult {
    /** Original text */
    original: string;
    /** Encoded phonetic representation */
    code: string;
}
/**
 * Tokenization result with sentence boundaries.
 */
interface TokenizationResult {
    /** Array of sentences */
    sentences: string[];
    /** Positions of sentence boundaries */
    positions: number[];
}
/**
 * Stemming result with removed affixes.
 */
interface StemmingResult {
    /** Original word */
    original: string;
    /** Stemmed word */
    stem: string;
    /** Removed affixes */
    removedAffixes: string[];
}

/**
 * ============================================================================
 * INDONESIAN NLP ENGINE - CONSTANTS
 * ============================================================================
 *
 * This file contains constants for Indonesian text processing:
 * - AFFIX_PATTERNS: Indonesian prefix and suffix patterns for stemming
 * - ABBREVIATIONS: Common Indonesian abbreviations to preserve in tokenization
 * - PHONETIC_MAP: Character mappings for phonetic encoding
 *
 * ============================================================================
 */
/**
 * Common Indonesian abbreviations that should NOT be split on periods
 * during sentence tokenization.
 */
declare const SENTENCE_ABBREVIATIONS: readonly ["Yth.", "Bpk.", "Ibu.", "Sdr.", "S.Kom.", "M.Kom.", "Dr.", "Sp.", "Mk.", "Jl.", "D.a.", "D.l."];
/**
 * Indonesian prefix patterns for algorithmic stemming.
 * Order matters - longer patterns should be checked first.
 */
declare const PREFIX_PATTERNS: readonly [{
    readonly pattern: RegExp;
    readonly replacement: "";
}, {
    readonly pattern: RegExp;
    readonly replacement: "s";
}, {
    readonly pattern: RegExp;
    readonly replacement: "";
}, {
    readonly pattern: RegExp;
    readonly replacement: "";
}, {
    readonly pattern: RegExp;
    readonly replacement: "";
}, {
    readonly pattern: RegExp;
    readonly replacement: "s";
}, {
    readonly pattern: RegExp;
    readonly replacement: "";
}, {
    readonly pattern: RegExp;
    readonly replacement: "";
}, {
    readonly pattern: RegExp;
    readonly replacement: "";
}, {
    readonly pattern: RegExp;
    readonly replacement: "";
}, {
    readonly pattern: RegExp;
    readonly replacement: "";
}, {
    readonly pattern: RegExp;
    readonly replacement: "";
}, {
    readonly pattern: RegExp;
    readonly replacement: "";
}, {
    readonly pattern: RegExp;
    readonly replacement: "";
}, {
    readonly pattern: RegExp;
    readonly replacement: "";
}, {
    readonly pattern: RegExp;
    readonly replacement: "";
}];
/**
 * Indonesian suffix patterns for algorithmic stemming.
 * Order matters - longer patterns should be checked first.
 */
declare const SUFFIX_PATTERNS: readonly [{
    readonly pattern: RegExp;
    readonly replacement: "";
}, {
    readonly pattern: RegExp;
    readonly replacement: "";
}, {
    readonly pattern: RegExp;
    readonly replacement: "";
}, {
    readonly pattern: RegExp;
    readonly replacement: "";
}, {
    readonly pattern: RegExp;
    readonly replacement: "";
}, {
    readonly pattern: RegExp;
    readonly replacement: "";
}, {
    readonly pattern: RegExp;
    readonly replacement: "";
}, {
    readonly pattern: RegExp;
    readonly replacement: "";
}, {
    readonly pattern: RegExp;
    readonly replacement: "";
}, {
    readonly pattern: RegExp;
    readonly replacement: "";
}, {
    readonly pattern: RegExp;
    readonly replacement: "";
}];

export { PREFIX_PATTERNS, type PhoneticResult, SENTENCE_ABBREVIATIONS, SUFFIX_PATTERNS, type StemmingResult, type TokenizationResult, encodePhonetic, isPhoneticMatch, normalizeWhitespace, stemText, stripNonAlphanumeric, tokenizeIndo };