<?php
/**
 * Template: robots-ai.txt (ADF-010) — AI crawler-specific directives.
 *
 * @package AIDF
 * @since   1.0.0
 *
 * @var array<string, mixed> $data Collected site data.
 */

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

// phpcs:disable WordPress.NamingConventions.PrefixAllGlobals.NonPrefixedVariableFound -- Template variables scoped to including method.

$identity    = $data['identity'];
$permissions = $data['permissions'];
$settings    = $data['settings'];

// Attribution header.
// phpcs:ignore WordPress.Security.EscapeOutput.OutputNotEscaped -- Pre-built internal string.
echo AIDF_Generator::get_attribution( '#', 'robots-ai-txt' );

echo '# AI Crawler Directives for ' . esc_html( $identity['name'] ) . "\n\n";

echo 'Website: [' . esc_html( $identity['url'] ) . '](' . esc_url( $identity['url'] ) . ')' . "\n";
echo 'Last Updated: ' . esc_html( wp_date( 'Y-m-d' ) ) . "\n\n";

echo "## Purpose\n\n";
echo "This file provides AI-specific crawler guidance that complements robots.txt.\n";
echo "It covers crawlers used by AI systems for training, retrieval, and indexing.\n\n";

// Known AI crawlers.
$crawlers = array(
	'GPTBot'           => 'OpenAI (ChatGPT, plugins)',
	'ChatGPT-User'     => 'OpenAI (ChatGPT browsing)',
	'ClaudeBot'        => 'Anthropic (Claude)',
	'Claude-Web'       => 'Anthropic (Claude web search)',
	'Google-Extended'  => 'Google (Gemini AI training)',
	'Googlebot'        => 'Google (Search, including AI Overviews)',
	'Bingbot'          => 'Microsoft (Bing, Copilot)',
	'PerplexityBot'    => 'Perplexity AI',
	'Bytespider'       => 'ByteDance (TikTok AI)',
	'CCBot'            => 'Common Crawl',
	'Applebot'         => 'Apple (Siri, Apple Intelligence)',
	'Meta-ExternalAgent' => 'Meta (AI training)',
);

echo "## Crawler Access\n\n";

switch ( $permissions['crawler_policy'] ) {
	case 'allow_all':
		echo "Policy: All AI crawlers are permitted.\n\n";
		foreach ( $crawlers as $name => $desc ) {
			echo esc_html( $name ) . ': Allow' . "\n";
			echo '  # ' . esc_html( $desc ) . "\n\n";
		}
		break;

	case 'allow_known':
		echo "Policy: Known AI crawlers are permitted for content retrieval.\n";
		echo "Training-specific crawlers may be restricted based on training policy.\n\n";

		foreach ( $crawlers as $name => $desc ) {
			$is_training = in_array( $name, array( 'Google-Extended', 'CCBot', 'Bytespider', 'Meta-ExternalAgent' ), true );

			if ( $is_training && 'allow' !== $permissions['ai_training'] ) {
				echo esc_html( $name ) . ': Disallow' . "\n";
				echo '  # ' . esc_html( $desc ) . ' — training crawler restricted per training policy' . "\n\n";
			} else {
				echo esc_html( $name ) . ': Allow' . "\n";
				echo '  # ' . esc_html( $desc ) . "\n\n";
			}
		}
		break;

	case 'restrict':
		echo "Policy: AI crawler access is restricted. Only select crawlers are permitted.\n\n";

		// Allow major search/AI assistants, block training-specific.
		$allowed = array( 'Googlebot', 'Bingbot', 'Applebot' );

		foreach ( $crawlers as $name => $desc ) {
			if ( in_array( $name, $allowed, true ) ) {
				echo esc_html( $name ) . ': Allow' . "\n";
				echo '  # ' . esc_html( $desc ) . "\n\n";
			} else {
				echo esc_html( $name ) . ': Disallow' . "\n";
				echo '  # ' . esc_html( $desc ) . "\n\n";
			}
		}
		break;
}

// Note about robots.txt.
echo "## Notes\n\n";
echo "- This file supplements robots.txt — it does not replace it.\n";
echo "- AI systems should also check ai.txt for content usage permissions.\n";
echo "- For structured permissions, see ai.json.\n";

$active = isset( $settings['active_files'] ) ? (array) $settings['active_files'] : array();

echo "\n## Related Files\n\n";
echo '- [robots.txt](' . esc_url( home_url( '/robots.txt' ) ) . ')' . "\n";

if ( in_array( 'ai-txt', $active, true ) ) {
	echo '- [ai.txt](' . esc_url( home_url( '/ai.txt' ) ) . ')' . "\n";
}

if ( in_array( 'ai-json', $active, true ) ) {
	echo '- [ai.json](' . esc_url( home_url( '/ai.json' ) ) . ')' . "\n";
}