<?php
/**
 * ARSCAN_Injector â€” Injects JSON-LD into <head> and filters robots.txt.
 *
 * Both hooks are server-side, so they work for AI crawlers that
 * do NOT execute JavaScript (GPTBot, ClaudeBot, PerplexityBot, etc.).
 */

defined( 'ABSPATH' ) || exit;

class ARSCAN_Injector {

    private static ?ARSCAN_Injector $instance = null;

    public static function get_instance(): ARSCAN_Injector {
        if ( null === self::$instance ) {
            self::$instance = new self();
        }
        return self::$instance;
    }

    private function __construct() {
        add_action( 'wp_head',      array( $this, 'inject_jsonld' ), 5 );
        add_filter( 'robots_txt',   array( $this, 'filter_robots_txt' ), 10, 2 );
    }

    // -----------------------------------------------------------------------
    // JSON-LD injection
    // -----------------------------------------------------------------------

    public function inject_jsonld(): void {
        $s = get_option( ARSCAN_OPTION_KEY, arscan_default_settings() );
        if ( empty( $s['inject_jsonld'] ) ) {
            return;
        }

        $schema = $this->build_schema( $s );
        if ( empty( $schema ) ) {
            return;
        }

        $json = wp_json_encode( $schema, JSON_PRETTY_PRINT | JSON_UNESCAPED_SLASHES | JSON_UNESCAPED_UNICODE );
        echo "\n<!-- AgentReadyScan JSON-LD -->\n";
        // phpcs:ignore WordPress.Security.EscapeOutput.OutputNotEscaped -- wp_json_encode output is safe for inline script content; esc_html would corrupt the JSON
        echo '<script type="application/ld+json">' . "\n" . $json . "\n" . '</script>' . "\n";
        echo "<!-- /AgentReadyScan -->\n\n";
    }

    private function build_schema( array $s ): array {
        $type = arscan_effective_schema_type();

        $schema = array(
            '@context' => 'https://schema.org',
            '@type'    => $type,
        );

        if ( ! empty( $s['business_name'] ) ) {
            $schema['name'] = $s['business_name'];
        }
        if ( ! empty( $s['business_url'] ) ) {
            $schema['url'] = $s['business_url'];
        }
        if ( ! empty( $s['description'] ) ) {
            $schema['description'] = $s['description'];
        }
        if ( ! empty( $s['phone'] ) ) {
            $schema['telephone'] = $s['phone'];
        }
        if ( ! empty( $s['email'] ) ) {
            $schema['email'] = $s['email'];
        }

        // Address
        $has_address = ! empty( $s['address'] ) || ! empty( $s['city'] ) || ! empty( $s['state'] );
        if ( $has_address ) {
            $postal_address = array( '@type' => 'PostalAddress' );
            if ( ! empty( $s['address'] ) ) $postal_address['streetAddress']   = $s['address'];
            if ( ! empty( $s['city'] ) )    $postal_address['addressLocality'] = $s['city'];
            if ( ! empty( $s['state'] ) )   $postal_address['addressRegion']   = $s['state'];
            if ( ! empty( $s['zip'] ) )     $postal_address['postalCode']      = $s['zip'];
            if ( ! empty( $s['country'] ) ) $postal_address['addressCountry']  = $s['country'];
            $schema['address'] = $postal_address;
        }

        // Opening hours
        if ( ! empty( $s['hours'] ) ) {
            $schema['openingHours'] = $s['hours'];
        }

        // Services
        if ( ! empty( $s['services'] ) ) {
            $service_list = array_filter( array_map( 'trim', explode( ',', $s['services'] ) ) );
            if ( ! empty( $service_list ) ) {
                // hasOfferCatalog is broadly applicable across schema types
                $schema['hasOfferCatalog'] = array(
                    '@type'           => 'OfferCatalog',
                    'name'            => 'Services',
                    'itemListElement' => array_values( array_map(
                        function( $svc ) {
                            return array(
                                '@type'       => 'Offer',
                                'itemOffered' => array(
                                    '@type' => 'Service',
                                    'name'  => $svc,
                                ),
                            );
                        },
                        $service_list
                    ) ),
                );
            }
        }

        // sameAs â€” llms.txt and agent-facts.html if they exist
        $same_as = array();
        if ( file_exists( ABSPATH . 'llms.txt' ) ) {
            $same_as[] = home_url( '/llms.txt' );
        }
        if ( file_exists( ABSPATH . 'agent-facts.html' ) ) {
            $same_as[] = home_url( '/agent-facts.html' );
        }
        if ( ! empty( $same_as ) ) {
            $schema['sameAs'] = count( $same_as ) === 1 ? $same_as[0] : $same_as;
        }

        return $schema;
    }

    // -----------------------------------------------------------------------
    // robots.txt filter
    // -----------------------------------------------------------------------

    public function filter_robots_txt( string $output, bool $public ): string {
        $s = get_option( ARSCAN_OPTION_KEY, arscan_default_settings() );
        if ( empty( $s['inject_robots'] ) ) {
            return $output;
        }

        $blocks = array(
            '',
            '# AI crawlers explicitly allowed',
            '# OpenAI',
            'User-agent: GPTBot',
            'Allow: /',
            'User-agent: ChatGPT-User',
            'Allow: /',
            'User-agent: OAI-SearchBot',
            'Allow: /',
            '',
            '# Anthropic',
            'User-agent: ClaudeBot',
            'Allow: /',
            'User-agent: anthropic-ai',
            'Allow: /',
            '',
            '# Perplexity',
            'User-agent: PerplexityBot',
            'Allow: /',
            '',
            '# Google AI training and retrieval',
            'User-agent: Google-Extended',
            'Allow: /',
            '',
            '# Meta',
            'User-agent: meta-externalagent',
            'Allow: /',
            '',
            '# Common Crawl',
            'User-agent: CCBot',
            'Allow: /',
            '',
            '# ByteDance',
            'User-agent: Bytespider',
            'Allow: /',
            '',
            '# Cohere',
            'User-agent: cohere-ai',
            'Allow: /',
            '',
            '# You.com',
            'User-agent: YouBot',
            'Allow: /',
            '',
            '# Apple',
            'User-agent: Applebot',
            'Allow: /',
            '',
            'Sitemap: ' . home_url( '/sitemap.xml' ),
        );

        if ( file_exists( ABSPATH . 'llms.txt' ) || file_exists( ABSPATH . 'agent-facts.html' ) ) {
            $blocks[] = '';
            $blocks[] = '# AI-facing public files';
            if ( file_exists( ABSPATH . 'llms.txt' ) ) {
                $blocks[] = '# ' . home_url( '/llms.txt' );
            }
            if ( file_exists( ABSPATH . 'agent-facts.html' ) ) {
                $blocks[] = '# ' . home_url( '/agent-facts.html' );
            }
        }

        $additions = implode( "\n", $blocks ) . "\n";

        return rtrim( $output ) . "\n" . $additions;
    }
}