<?php
/**
 * Pure validation helpers (no WordPress dependency, unit-testable).
 *
 * @package Soft_ML
 */

defined( 'ABSPATH' ) || exit;

class Soft_ML_Validator {

	/** URL segments that must never be used as a language slug. */
	const RESERVED_SLUGS = array(
		'wp-admin', 'wp-content', 'wp-includes', 'wp-json', 'wp-login', 'wp-login.php',
		'feed', 'feeds', 'rss', 'rss2', 'atom', 'comments', 'page', 'search', 'author',
		'category', 'tag', 'embed', 'trackback', 'attachment', 'amp', 'admin', 'login',
		'sitemap', 'sitemap.xml', 'sitemap_index.xml', 'robots.txt', 'xmlrpc.php',
	);

	/**
	 * Language code: 2-3 letters, optional region/script subtag (en, fr, pt-br, zh-hans).
	 *
	 * @param string $code Code.
	 * @return bool
	 */
	public static function is_valid_code( $code ) {
		return is_string( $code ) && 1 === preg_match( '/^[a-z]{2,3}(-[a-z0-9]{2,8})?$/', $code );
	}

	/**
	 * URL slug for a language directory.
	 *
	 * @param string $slug Slug.
	 * @return bool
	 */
	public static function is_valid_slug( $slug ) {
		return is_string( $slug )
			&& 1 === preg_match( '/^[a-z0-9]+(-[a-z0-9]+)*$/', $slug )
			&& ! in_array( $slug, self::RESERVED_SLUGS, true );
	}

	/**
	 * WordPress-style locale (en_US, fr_FR, pt_BR).
	 *
	 * @param string $locale Locale.
	 * @return bool
	 */
	public static function is_valid_locale( $locale ) {
		return is_string( $locale ) && 1 === preg_match( '/^[a-z]{2,3}(_[A-Z]{2})?$/', $locale );
	}

	/**
	 * ISO 3166-1 alpha-2 country code, uppercase.
	 *
	 * @param string $cc Country code.
	 * @return bool
	 */
	public static function is_valid_country( $cc ) {
		return is_string( $cc ) && 1 === preg_match( '/^[A-Z]{2}$/', $cc );
	}

	/**
	 * Parse "US=en" lines (one per line) into [ 'US' => 'en' ]. Invalid lines are skipped.
	 *
	 * @param string $text Raw text.
	 * @return array
	 */
	public static function parse_country_map( $text ) {
		$map = array();
		foreach ( preg_split( '/\R/', (string) $text ) as $line ) {
			$parts = array_map( 'trim', explode( '=', $line, 2 ) );
			if ( 2 !== count( $parts ) ) {
				continue;
			}
			$cc   = strtoupper( $parts[0] );
			$lang = strtolower( $parts[1] );
			if ( self::is_valid_country( $cc ) && self::is_valid_code( $lang ) ) {
				$map[ $cc ] = $lang;
			}
		}
		return $map;
	}
}