<?php
/**
 * Text utility helper.
 *
 * @package AISupportAssistant\Helpers
 */

namespace AISupportAssistant\Helpers;

if ( ! defined( 'ABSPATH' ) ) {
	exit;
}

/**
 * Class TextHelper
 */
class TextHelper {

	/**
	 * Clean and normalize extracted text.
	 *
	 * @param string $text Raw text.
	 * @return string Cleaned text.
	 */
	public static function clean_text( string $text ): string {
		// Convert Smart Quotes and dashes to standard characters.
		$smart_quotes = array(
			"\xe2\x80\x98" => "'", // Left single quote
			"\xe2\x80\x99" => "'", // Right single quote
			"\xe2\x80\x9c" => '"', // Left double quote
			"\xe2\x80\x9d" => '"', // Right double quote
			"\xe2\x80\x93" => '-', // En dash
			"\xe2\x80\x94" => '-', // Em dash
		);
		$text         = strtr( $text, $smart_quotes );

		// Decode HTML entities if any.
		$text = html_entity_decode( $text, ENT_QUOTES | ENT_HTML5, 'UTF-8' );

		// Remove control characters except newlines/tabs.
		$text = preg_replace( '/[^\PC\s]/u', '', $text );

		// Normalize line endings to LF.
		$text = str_replace( array( "\r\n", "\r" ), "\n", $text );

		// Replace multiple spaces with a single space.
		$text = preg_replace( '/[ \t]+/u', ' ', $text );

		// Replaces consecutive newlines with maximum two newlines (to preserve paragraphs).
		$text = preg_replace( '/\n{3,}/u', "\n\n", $text );

		return trim( $text );
	}

	/**
	 * Approximate the token count of a string.
	 * Handled via a hybrid calculation of characters and words:
	 * (1 word = 1.33 tokens, 4 chars = 1 token).
	 *
	 * @param string $text Text to token count.
	 * @return int Estimated token count.
	 */
	public static function count_tokens( string $text ): int {
		if ( empty( $text ) ) {
			return 0;
		}

		$word_count = str_word_count( $text );
		$char_count = mb_strlen( $text, 'UTF-8' );

		$token_by_words = (int) ceil( $word_count / 0.75 );
		$token_by_chars = (int) ceil( $char_count / 4.0 );

		// Return the max of the two estimates to avoid underestimating token usage limits.
		return max( $token_by_words, $token_by_chars );
	}

	/**
	 * Split text into sentences using standard regex.
	 *
	 * @param string $text Text to split.
	 * @return array Array of sentences.
	 */
	public static function split_into_sentences( string $text ): array {
		if ( empty( $text ) ) {
			return array();
		}

		// Regex pattern to split sentences without stripping punctuation.
		// Splits on '.', '!', or '?' followed by whitespace, keeping the punctuation attached.
		$sentences = preg_split( '/(?<=[.!?])\s+/u', $text, -1, PREG_SPLIT_NO_EMPTY );

		if ( ! is_array( $sentences ) ) {
			return array( $text );
		}

		return array_map( 'trim', $sentences );
	}
}