<?php
/**
 * LLMS.txt Content Generator
 * 
 * Generates llms.txt content following the specification at:
 * https://github.com/AnswerDotAI/llms-txt
 * 
 * @package AI_Ready_Core
 * @since 1.0.0
 */

// Prevent direct access
if (!defined('ABSPATH')) {
    exit;
}

/**
 * Initialize default options for llms.txt generation
 */
function ai_ready_init_llms_defaults() {
    // Set default summary if not exists
    if (get_option('ai_ready_llms_summary') === false) {
        $default_summary = get_bloginfo('description');
        if (empty($default_summary)) {
            $default_summary = __('A WordPress website providing information and resources.', 'ai-ready-core');
        }
        add_option('ai_ready_llms_summary', $default_summary);
    }
    
    // Set default sections if not exists (basic fallback)
    if (get_option('ai_ready_llms_sections') === false) {
        // Use the comprehensive default sections builder
        $default_sections = ai_ready_get_default_sections();
        add_option('ai_ready_llms_sections', $default_sections);
    }
    
    // Initialize notes as empty
    if (get_option('ai_ready_llms_notes') === false) {
        add_option('ai_ready_llms_notes', '');
    }
}

/**
 * Generate llms.txt content - ENHANCED VERSION
 */
function ai_ready_generate_llms_txt_enhanced() {
    try {
        if (!ai_ready_is_llms_enabled()) {
            return '';
        }
        
        $last_refresh = get_option('ai_ready_last_section_refresh', 0);
        
        $current_time = time();
        
        $time_diff = $current_time - $last_refresh;
        
        $force_refresh = $time_diff > DAY_IN_SECONDS;
        
        if ($force_refresh) {
            ai_ready_force_refresh_sections();
            
            update_option('ai_ready_last_section_refresh', time());
        }
        
        $custom_content = get_option('ai_ready_custom_content', '');
        
        if (!empty($custom_content)) {
            return apply_filters('ai_ready_llms_txt_content', $custom_content);
        }
        
        $lines = [];
        
        $site_title_option = get_option('ai_ready_site_title', false);
        
        if ($site_title_option === false || empty($site_title_option)) {
            $site_title_from_bloginfo = get_bloginfo('name');
            $site_title = $site_title_from_bloginfo;
        } else {
            $site_title = $site_title_option;
        }
        
        $site_title = is_string($site_title) ? $site_title : (string)$site_title;
        
        $lines[] = "# {$site_title}";
        $lines[] = "";
        
        $site_summary = get_option('ai_ready_site_summary', '');
        
        if (empty($site_summary)) {
            $site_summary = get_option('ai_ready_llms_summary', '');
        }
        if (empty($site_summary)) {
            $site_summary = get_bloginfo('description');
        }
        
        if (!empty($site_summary)) {
            $lines[] = "> {$site_summary}";
            $lines[] = "";
        }
        
        $additional_notes = get_option('ai_ready_llms_notes', '');
        
        if (empty($additional_notes)) {
            $auto_notes = ai_ready_generate_auto_notes();
            
            if (!empty($auto_notes)) {
                $lines[] = trim($auto_notes);
                $lines[] = "";
            }
        } else {
            $lines[] = trim($additional_notes);
            $lines[] = "";
        }
        
        ai_ready_init_llms_defaults();
        
        $sections = get_option('ai_ready_llms_sections', []);
        
        $needs_upgrade = false;
        if (empty($sections) || $force_refresh) {
            $needs_upgrade = true;
        } elseif (count($sections) === 1 && isset($sections[0]['name']) && strtolower($sections[0]['name']) === 'docs') {
            $links = $sections[0]['links'] ?? [];
            if (count($links) <= 1) {
                $needs_upgrade = true;
            }
        }
        
        if ($needs_upgrade) {
            $sections = ai_ready_get_default_sections();
            
            update_option('ai_ready_llms_sections', $sections);
        }
        
        foreach ($sections as $section) {
            if (empty($section['name']) || empty($section['links'])) {
                continue;
            }
            
            $section_name = isset($section['name']) ? (string)$section['name'] : '';
            $lines[] = "## {$section_name}";
            $lines[] = "";
            
            foreach ($section['links'] as $link) {
                if (empty($link['title']) || empty($link['url'])) {
                    continue;
                }
                
                $link_title = isset($link['title']) ? (string)$link['title'] : '';
                $link_url = isset($link['url']) ? (string)$link['url'] : '';
                $link_description = isset($link['description']) ? (string)$link['description'] : '';
                
                $description = !empty($link_description) ? ": {$link_description}" : '';
                $lines[] = "- [{$link_title}]({$link_url}){$description}";
            }
            
            $lines[] = "";
        }
        
        $content = implode("\n", $lines);
        
        return apply_filters('ai_ready_llms_txt_content', $content);
        
    } catch (Exception $e) {
        return "# " . __('Error', 'ai-ready-core') . "\n\n" . __('An error occurred while generating llms.txt content.', 'ai-ready-core');
    }
}

/**
 * Get default sections for llms.txt
 * 
 * Creates sensible default sections when no custom sections are configured
 * Automatically detects and includes relevant content following the llms.txt spec
 * 
 * @return array Default sections array
 */
if (!function_exists('ai_ready_get_default_sections')) {
    function ai_ready_get_default_sections() {
        $sections = [];
        
        $direct_pages = ai_ready_get_pages_direct();
        
        if (!empty($direct_pages)) {
            $categorized_pages = ai_ready_categorize_sitemap_pages($direct_pages);
            
            foreach ($categorized_pages as $category => $pages) {
                if (empty($pages)) continue;
                
                $section_name = ucfirst($category);
                $section_links = [];
                
                if ($category === 'docs') {
                    $section_links[] = [
                        'title' => __('Home', 'ai-ready-core'),
                        'url' => home_url(),
                        'description' => __('Main website homepage', 'ai-ready-core')
                    ];
                }
                
                foreach ($pages as $page) {
                    $description = ai_ready_generate_page_description($page, $category);
                    $section_links[] = [
                        'title' => $page['title'],
                        'url' => $page['url'],
                        'description' => $description
                    ];
                }
                
                if (!empty($section_links)) {
                    $sections[] = [
                        'name' => $section_name,
                        'links' => $section_links
                    ];
                }
            }
            
            if (isset($categorized_pages['optional'])) {
                $optional_index = array_search('Optional', array_column($sections, 'name'));
                if ($optional_index !== false) {
                    array_unshift($sections[$optional_index]['links'], 
                        [
                            'title' => __('Sitemap', 'ai-ready-core'),
                            'url' => home_url('/sitemap.xml'),
                            'description' => __('Complete site structure for navigation', 'ai-ready-core')
                        ],
                        [
                            'title' => __('RSS Feed', 'ai-ready-core'),
                            'url' => get_feed_link(),
                            'description' => __('Syndicated content updates', 'ai-ready-core')
                        ]
                    );
                }
            }
            
        } else {
            ai_ready_refresh_sitemap();
            
            $sitemap_pages = ai_ready_get_pages_from_sitemap();
            
            if (!empty($sitemap_pages)) {
                $categorized_pages = ai_ready_categorize_sitemap_pages($sitemap_pages);
            } else {
                $sections = ai_ready_get_default_sections_manual();
            }
        }
        
        return $sections;
    }
}

/**
 * Get pages directly from WordPress database (most reliable method)
 * 
 * @return array Array of page data
 */
function ai_ready_get_pages_direct() {
    $pages = [];
    
    $wp_pages = get_pages([
        'post_status' => 'publish',
        'number' => 100,
        'sort_column' => 'menu_order, post_title'
    ]);
    
    foreach ($wp_pages as $page) {
        $page_data = [
            'id' => $page->ID,
            'title' => $page->post_title,
            'url' => get_permalink($page),
            'type' => 'page',
            'slug' => $page->post_name,
            'content_length' => strlen(wp_strip_all_tags($page->post_content))
        ];
        $pages[] = $page_data;
    }
    
    $wp_posts = get_posts([
        'post_status' => 'publish',
        'numberposts' => 10,
        'post_type' => 'post'
    ]);
    
    foreach ($wp_posts as $post) {
        $content_length = strlen(wp_strip_all_tags($post->post_content));
        
        if ($content_length > 500) {
            $pages[] = [
                'id' => $post->ID,
                'title' => $post->post_title,
                'url' => get_permalink($post),
                'type' => 'post',
                'slug' => $post->post_name,
                'content_length' => $content_length
            ];
        }
    }
    
    return $pages;
}

/**
 * Refresh WordPress sitemaps to ensure they're up to date
 */
function ai_ready_refresh_sitemap() {
    if (function_exists('wp_sitemaps_get_server')) {
        $sitemaps = wp_sitemaps_get_server();
        if ($sitemaps && method_exists($sitemaps, 'registry')) {
            wp_cache_flush();
        }
    }
    
    if (function_exists('wpseo_invalidate_sitemap_cache')) {
        wpseo_invalidate_sitemap_cache();
    }
    
    if (class_exists('RankMath\Sitemap\Cache_Watcher')) {
        do_action('rank_math/sitemap/invalidate_cache');
    }
    
    if (function_exists('aioseo')) {
        if (method_exists(aioseo()->sitemap, 'clearCache')) {
            aioseo()->sitemap->clearCache();
        }
    }
}

/**
 * Generate appropriate description for a page based on its category
 * 
 * @param array $page Page data
 * @param string $category Page category
 * @return string Description text
 */
function ai_ready_generate_page_description($page, $category) {
    $slug = strtolower($page['slug']);
    $title = strtolower($page['title']);
    
    if (strpos($slug, 'about') !== false || strpos($title, 'about') !== false) {
        return __('Information about this organization', 'ai-ready-core');
    }
    
    if (strpos($slug, 'mission') !== false || strpos($title, 'mission') !== false) {
        return __('Organization mission and vision', 'ai-ready-core');
    }
    
    if (strpos($slug, 'service') !== false || strpos($title, 'service') !== false) {
        return __('Services and products offered', 'ai-ready-core');
    }

    if (strpos($slug, 'products') !== false || strpos($title, 'products') !== false) {
        return __('Products offered', 'ai-ready-core');
    }
    
    if (strpos($slug, 'contact') !== false || strpos($title, 'contact') !== false) {
        return __('Contact information and methods', 'ai-ready-core');
    }
    
    if (strpos($slug, 'faq') !== false || strpos($title, 'faq') !== false) {
        return __('Frequently asked questions and support', 'ai-ready-core');
    }
    
    if (strpos($slug, 'portfolio') !== false || strpos($title, 'portfolio') !== false) {
        return 'Examples of work and projects';
    }
    
    if (strpos($slug, 'resource') !== false || strpos($title, 'resource') !== false) {
        return 'Additional resources and documentation';
    }
    
    if (strpos($slug, 'privacy') !== false || strpos($title, 'privacy') !== false) {
        return 'Privacy policy and data handling practices';
    }
    
    if (strpos($slug, 'terms') !== false || strpos($title, 'terms') !== false) {
        return 'Terms of service and usage conditions';
    }
    
    $extracted_description = ai_ready_extract_page_description($page);
    if ($extracted_description) {
        return $extracted_description;
    }
    
    switch ($category) {
        case 'docs':
            return 'Important site information';
        case 'content':
            return $page['content_length'] > 1000 ? 'Detailed content and information' : 'Important content';
        case 'resources':
            return 'Resources and additional materials';
        case 'optional':
            return 'Additional information and resources';
        default:
            return 'Website content';
    }
}

/**
 * Extract meaningful description from page content
 * 
 * @param array $page Page data with ID
 * @return string|false Extracted description or false if none found
 */
function ai_ready_extract_page_description($page) {
    if (empty($page['id'])) {
        return false;
    }
    
    $post = get_post($page['id']);
    if (!$post) {
        return false;
    }
    
    if (!empty($post->post_excerpt)) {
        $excerpt = wp_strip_all_tags($post->post_excerpt);
        $excerpt = preg_replace('/\s+/', ' ', $excerpt);
        $excerpt = trim($excerpt);
        
        if (strlen($excerpt) > 20 && strlen($excerpt) <= 150) {
            return $excerpt;
        } elseif (strlen($excerpt) > 150) {
            return substr($excerpt, 0, 147) . '...';
        }
    }
    
    $content = $post->post_content;
    if (empty($content)) {
        return false;
    }
    
    $content = do_shortcode($content);
    $content = wp_strip_all_tags($content);
    $content = preg_replace('/\s+/', ' ', $content);
    $content = trim($content);
    
    if (empty($content)) {
        return false;
    }
    
    $first_sentence = ai_ready_extract_first_sentence($content);
    if ($first_sentence && strlen($first_sentence) >= 20 && strlen($first_sentence) <= 200) {
        return $first_sentence;
    }
    
    if (strlen($content) > 120) {
        $truncated = substr($content, 0, 117);
        $last_space = strrpos($truncated, ' ');
        if ($last_space > 80) {
            $truncated = substr($truncated, 0, $last_space);
        }
        return $truncated . '...';
    } elseif (strlen($content) >= 20) {
        return $content;
    }
    
    return false;
}

/**
 * Extract the first meaningful sentence from text
 * 
 * @param string $text Content text
 * @return string|false First sentence or false if none found
 */
function ai_ready_extract_first_sentence($text) {
    $sentences = preg_split('/[.!?]+/', $text, 3);
    
    if (!empty($sentences[0])) {
        $first_sentence = trim($sentences[0]);
        
        if (strlen($first_sentence) >= 20 && str_word_count($first_sentence) >= 4) {
            return $first_sentence;
        }
        
        if (!empty($sentences[1]) && strlen($first_sentence) < 60) {
            $combined = trim($first_sentence . '. ' . trim($sentences[1]));
            if (strlen($combined) <= 200) {
                return $combined;
            }
        }
    }
    
    return false;
}

/**
 * Fallback manual page discovery (original method)
 */
function ai_ready_get_default_sections_manual() {
    $sections = [];
    
    $docs_links = [];
    
    $docs_links[] = [
        'title' => 'Home',
        'url' => home_url(),
        'description' => 'Main website homepage'
    ];
    
    $find_page = function($variants, $title_variants = []) {
        foreach ($variants as $variant) {
            $page = get_page_by_path($variant);
            if ($page) {
                return $page;
            }
        }
        
        if (!empty($title_variants)) {
            $pages = get_pages([
                'post_status' => 'publish',
                'number' => 10,
                'sort_column' => 'post_title'
            ]);
            
            foreach ($pages as $page) {
                $page_title = strtolower($page->post_title);
                foreach ($title_variants as $variant) {
                    if (strpos($page_title, strtolower($variant)) !== false) {
                        return $page;
                    }
                }
            }
        }
        
        return null;
    };
        
        $about_page = $find_page(
            ['about', 'about-us', 'who-we-are', 'our-story', 'company', 'mission', 'vision'],
            ['about', 'who we are', 'our story', 'mission', 'vision']
        );
        if ($about_page) {
            $docs_links[] = [
                'title' => $about_page->post_title,
                'url' => get_permalink($about_page),
                'description' => 'Information about this organization'
            ];
        }
        
        $mission_page = $find_page(
            ['mission', 'vision', 'mission-vision', 'mission-statement', 'our-mission'],
            ['mission', 'vision', 'mission statement', 'our mission']
        );
        if ($mission_page && (!$about_page || $mission_page->ID !== $about_page->ID)) {
            $docs_links[] = [
                'title' => $mission_page->post_title,
                'url' => get_permalink($mission_page),
                'description' => 'Organization mission and vision'
            ];
        }
        
        $service_page = $find_page(
            ['services', 'products', 'offerings', 'solutions', 'what-we-do', 'capabilities'],
            ['services', 'products', 'offerings', 'solutions', 'what we do', 'capabilities']
        );
        if ($service_page) {
            $docs_links[] = [
                'title' => $service_page->post_title,
                'url' => get_permalink($service_page),
                'description' => 'Services and products offered'
            ];
        }
        
        $faq_page = $find_page(
            ['faq', 'help', 'support', 'frequently-asked-questions', 'knowledge-base', 'faqs'],
            ['faq', 'faqs', 'help', 'support', 'frequently asked questions', 'knowledge base']
        );
        if ($faq_page) {
            $docs_links[] = [
                'title' => $faq_page->post_title,
                'url' => get_permalink($faq_page),
                'description' => 'Frequently asked questions and support'
            ];
        }
        
        $contact_page = $find_page(
            ['contact', 'contact-us', 'get-in-touch', 'reach-us', 'connect'],
            ['contact', 'contact us', 'get in touch', 'reach us', 'connect']
        );
        if ($contact_page) {
            $docs_links[] = [
                'title' => $contact_page->post_title,
                'url' => get_permalink($contact_page),
                'description' => 'Contact information and methods'
            ];
        }
        
        if (!empty($docs_links)) {
            $sections[] = [
                'name' => 'Docs',
                'links' => $docs_links
            ];
        }
        
        $optional_links = [];
        
        $optional_links[] = [
            'title' => 'Sitemap',
            'url' => home_url('/sitemap.xml'),
            'description' => 'Complete site structure for navigation'
        ];
        
        $optional_links[] = [
            'title' => 'RSS Feed',
            'url' => get_feed_link(),
            'description' => 'Syndicated content updates'
        ];
        
        $privacy_page_id = get_option('wp_page_for_privacy_policy');
        if ($privacy_page_id) {
            $optional_links[] = [
                'title' => get_the_title($privacy_page_id),
                'url' => get_permalink($privacy_page_id),
                'description' => 'Data handling and privacy practices'
            ];
        }
        
        if (!empty($optional_links)) {
            $sections[] = [
                'name' => 'Optional',
                'links' => $optional_links
            ];
        }
        
        return $sections;
    }

/**
 * Get social media links from theme customizer or common sources
 * 
 * @return array Array of social media links
 */
function ai_ready_get_social_links() {
    $social_links = [];
    
    $social_options = [
        'facebook' => 'Facebook',
        'twitter' => 'Twitter', 
        'linkedin' => 'LinkedIn',
        'instagram' => 'Instagram',
        'youtube' => 'YouTube',
        'github' => 'GitHub'
    ];
    
    foreach ($social_options as $platform => $label) {
        $url = get_theme_mod($platform . '_url') ?: 
               get_theme_mod('social_' . $platform) ?: 
               get_option($platform . '_url') ?: 
               get_option('social_' . $platform);
        
        if ($url && filter_var($url, FILTER_VALIDATE_URL)) {
            $social_links[] = [
                'title' => $label,
                'url' => $url,
                'description' => 'Follow on ' . $label
            ];
        }
    }
    
    return $social_links;
}

/**
 * Generate automatic contextual notes for llms.txt
 * 
 * @return string Auto-generated notes
 */
if (!function_exists('ai_ready_generate_auto_notes')) {
function ai_ready_generate_auto_notes() {
    $notes = [];
    
    try {
        $site_url = home_url();
        $admin_email = get_option('admin_email');
        
        $post_count_obj = wp_count_posts('post');
        $post_count = (is_object($post_count_obj) && isset($post_count_obj->publish)) ? (int)$post_count_obj->publish : 0;
        
        $page_count_obj = wp_count_posts('page');
        $page_count = (is_object($page_count_obj) && isset($page_count_obj->publish)) ? (int)$page_count_obj->publish : 0;
        
        if ($post_count > 0 && $page_count > 0) {
            $notes[] = sprintf("This website contains %d published articles and %d static pages.", $post_count, $page_count);
        } elseif ($post_count > 0) {
            $notes[] = sprintf("This is a blog-focused website with %d published articles.", $post_count);
        } elseif ($page_count > 0) {
            $notes[] = sprintf("This is a static website with %d main pages.", $page_count);
        }
        
        $active_plugins = get_option('active_plugins', []);
        if (!is_array($active_plugins)) {
            $active_plugins = [];
        }
        $features = [];
        
        if (get_option('users_can_register')) {
            $features[] = "user registration";
        }
        
        if (get_option('default_comment_status') === 'open') {
            $features[] = "commenting system";
        }
        
        if (in_array('woocommerce/woocommerce.php', $active_plugins) || class_exists('WooCommerce')) {
            $features[] = "e-commerce functionality";
        }
        
        if (in_array('polylang/polylang.php', $active_plugins) || 
            in_array('sitepress-multilingual-cms/sitepress.php', $active_plugins)) {
            $features[] = "multilingual content";
        }
        
        if (in_array('learnpress/learnpress.php', $active_plugins) || 
            in_array('learndash/learndash.php', $active_plugins)) {
            $features[] = "learning management system";
        }
        
        if (!empty($features)) {
            $notes[] = "Key features: " . implode(', ', $features) . ".";
        }
        
        if (function_exists('wp_get_theme')) {
            $theme = wp_get_theme();
            $theme_name = '';
            
            if (is_object($theme) && method_exists($theme, 'get')) {
                $theme_name = $theme->get('Name');
                if (!is_string($theme_name)) {
                    $theme_name = '';
                }
            }
            
            if (!empty($theme_name)) {
                $theme_name_lower = strtolower($theme_name);
                if (strpos($theme_name_lower, 'business') !== false || 
                    strpos($theme_name_lower, 'corporate') !== false) {
                    $notes[] = "This appears to be a business or corporate website.";
                } elseif (strpos($theme_name_lower, 'portfolio') !== false) {
                    $notes[] = "This appears to be a portfolio website showcasing work and projects.";
                } elseif ($post_count > $page_count * 2) {
                    $notes[] = "This is primarily a content-driven site with regular blog updates.";
                }
            }
        }
        
        if (function_exists('get_posts')) {
            $recent_post = get_posts(['numberposts' => 1, 'post_status' => 'publish']);
            if (!empty($recent_post) && is_array($recent_post) && isset($recent_post[0])) {
                $post = $recent_post[0];
                if (is_object($post) && isset($post->post_date)) {
                    $last_updated = strtotime($post->post_date);
                    if ($last_updated !== false) {
                        $days_ago = floor((time() - $last_updated) / (60 * 60 * 24));
                        
                        if ($days_ago < 7) {
                            $notes[] = "Content is updated regularly (last updated within the past week).";
                        } elseif ($days_ago < 30) {
                            $notes[] = "Content is updated periodically (last updated within the past month).";
                        }
                    }
                }
            }
        }
        
    } catch (Exception $e) {
        return "This is a WordPress website.";
    }
    
    return implode(" ", $notes);
}
}

/**
 * Get available pages for section building
 * 
 * Helper function to get a list of published pages that could be included
 * in llms.txt sections. Useful for admin interface.
 * 
 * @return array Array of page objects with ID, title, and URL
 */
function ai_ready_get_available_pages() {
    $pages = get_pages([
        'post_status' => 'publish',
        'number' => 50,
        'sort_column' => 'menu_order, post_title'
    ]);
    
    $available_pages = [];
    
    foreach ($pages as $page) {
        $available_pages[] = [
            'id' => $page->ID,
            'title' => $page->post_title,
            'url' => get_permalink($page),
            'slug' => $page->post_name
        ];
    }
    
    return $available_pages;
}

/**
 * Validate llms.txt content format
 * 
 * Checks if the generated content follows the llms.txt specification
 * 
 * @param string $content The llms.txt content to validate
 * @return array Validation results with 'valid' boolean and 'errors' array
 */
function ai_ready_validate_llms_txt($content) {
    $errors = [];
    $lines = explode("\n", $content);
    
    $first_line = '';
    foreach ($lines as $line) {
        if (!empty(trim($line))) {
            $first_line = trim($line);
            break;
        }
    }
    
    if (!preg_match('/^# .+/', $first_line)) {
        $errors[] = 'Content must start with an H1 title (# Title)';
    }
    
    $has_section = false;
    foreach ($lines as $line) {
        if (preg_match('/^## .+/', trim($line))) {
            $has_section = true;
            break;
        }
    }
    
    if (!$has_section) {
        $errors[] = 'Content must have at least one section (## Section Name)';
    }
    
    $in_section = false;
    foreach ($lines as $line) {
        $trimmed = trim($line);
        
        if (preg_match('/^## .+/', $trimmed)) {
            $in_section = true;
            continue;
        }
        
        if ($in_section && preg_match('/^- /', $trimmed)) {
            if (!preg_match('/^- \[.+\]\(.+\)/', $trimmed)) {
                $errors[] = 'Section links must use markdown format: - [Title](URL)';
                break;
            }
        }
    }
    
    return [
        'valid' => empty($errors),
        'errors' => $errors
    ];
}

add_action('ai_ready_core_activated', 'ai_ready_init_llms_defaults');

/**
 * Public wrapper matching spec name.
 * Delegates to the enhanced generator.
 *
 * @return string
 */
function ai_ready_generate_llms_txt() {
    return ai_ready_generate_llms_txt_enhanced();
}

/**
 * Get pages from sitemap for comprehensive page discovery
 * 
 * @return array Array of page objects with URL, title, and type
 */
function ai_ready_get_pages_from_sitemap() {
    $pages = [];
    
    $sitemap_url = home_url('/sitemap.xml');
    
    $sitemap_urls = [
        home_url('/wp-sitemap.xml'),
        home_url('/sitemap.xml'),
        home_url('/sitemap_index.xml'),
        home_url('/post-sitemap.xml'),
        home_url('/page-sitemap.xml'),
    ];
    
    foreach ($sitemap_urls as $url) {
        $sitemap_content = ai_ready_fetch_sitemap($url);
        if ($sitemap_content) {
            $discovered_pages = ai_ready_parse_sitemap($sitemap_content);
            $pages = array_merge($pages, $discovered_pages);
        }
    }
    
    $unique_pages = [];
    $seen_urls = [];
    
    foreach ($pages as $page) {
        if (!in_array($page['url'], $seen_urls)) {
            $unique_pages[] = $page;
            $seen_urls[] = $page['url'];
        }
    }
    
    return $unique_pages;
}

/**
 * Fetch sitemap content
 * 
 * @param string $url Sitemap URL
 * @return string|false Sitemap content or false on failure
 */
function ai_ready_fetch_sitemap($url) {
    $response = wp_remote_get($url, [
        'timeout' => 10,
        'headers' => [
            'User-Agent' => 'AI-Ready-Plugin/1.0'
        ]
    ]);
    
    if (!is_wp_error($response) && wp_remote_retrieve_response_code($response) === 200) {
        return wp_remote_retrieve_body($response);
    }
    
    if (strpos($url, '/wp-sitemap.xml') !== false) {
        $sitemap_content = ai_ready_generate_local_sitemap();
        if ($sitemap_content) {
            return $sitemap_content;
        }
    }
    
    return false;
}

/**
 * Generate a basic sitemap from WordPress data
 * 
 * @return string Simple XML sitemap
 */
function ai_ready_generate_local_sitemap() {
    $pages = get_pages([
        'post_status' => 'publish',
        'number' => 100
    ]);
    
    $posts = get_posts([
        'post_status' => 'publish',
        'numberposts' => 50,
        'post_type' => 'post'
    ]);
    
    $xml = '<?xml version="1.0" encoding="UTF-8"?>' . "\n";
    $xml .= '<urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9">' . "\n";
    
    $xml .= '<url><loc>' . home_url() . '</loc></url>' . "\n";
    
    foreach ($pages as $page) {
        $xml .= '<url><loc>' . get_permalink($page) . '</loc></url>' . "\n";
    }
    
    foreach ($posts as $post) {
        $xml .= '<url><loc>' . get_permalink($post) . '</loc></url>' . "\n";
    }
    
    $xml .= '</urlset>';
    
    return $xml;
}

/**
 * Parse sitemap XML content
 * 
 * @param string $xml_content Sitemap XML content
 * @return array Array of discovered pages
 */
function ai_ready_parse_sitemap($xml_content) {
    $pages = [];
    
    if (strpos($xml_content, '<sitemapindex') !== false) {
        $pages = ai_ready_parse_sitemap_index($xml_content);
    } else {
        $pages = ai_ready_parse_sitemap_urlset($xml_content);
    }
    
    return $pages;
}

/**
 * Parse sitemap index (that contains links to other sitemaps)
 * 
 * @param string $xml_content Sitemap index XML
 * @return array Array of pages from all linked sitemaps
 */
function ai_ready_parse_sitemap_index($xml_content) {
    $pages = [];
    
    preg_match_all('/<loc>(.*?)<\/loc>/', $xml_content, $matches);
    
    if (!empty($matches[1])) {
        foreach ($matches[1] as $sitemap_url) {
            if (strpos($sitemap_url, 'page-sitemap') !== false || 
                strpos($sitemap_url, 'wp-sitemap-pages') !== false) {
                $sub_sitemap = ai_ready_fetch_sitemap($sitemap_url);
                if ($sub_sitemap) {
                    $sub_pages = ai_ready_parse_sitemap_urlset($sub_sitemap);
                    $pages = array_merge($pages, $sub_pages);
                }
            }
        }
    }
    
    return $pages;
}

/**
 * Parse sitemap urlset (actual page URLs)
 * 
 * @param string $xml_content Sitemap XML with urlset
 * @return array Array of page data
 */
function ai_ready_parse_sitemap_urlset($xml_content) {
    $pages = [];
    
    preg_match_all('/<loc>(.*?)<\/loc>/', $xml_content, $matches);
    
    if (!empty($matches[1])) {
        foreach ($matches[1] as $url) {
            if (ai_ready_is_relevant_page_url($url)) {
                $page_data = ai_ready_get_page_info_from_url($url);
                if ($page_data) {
                    $pages[] = $page_data;
                }
            }
        }
    }
    
    return $pages;
}

/**
 * Check if a URL represents a relevant page for llms.txt
 * 
 * @param string $url Page URL
 * @return bool True if relevant
 */
function ai_ready_is_relevant_page_url($url) {
    $home_url = trailingslashit(home_url());
    
    if (strpos($url, $home_url) !== 0) {
        return false;
    }
    
    $skip_patterns = [
        '/wp-admin/',
        '/wp-content/',
        '/feed/',
        '/comment/',
        '/trackback/',
        '/xmlrpc.php',
        '?',
        '#',
        '/category/',
        '/tag/',
        '/author/',
        '/date/',
        '/search/',
        '/attachment/'
    ];
    
    foreach ($skip_patterns as $pattern) {
        if (strpos($url, $pattern) !== false) {
            return false;
        }
    }
    
    return true;
}

/**
 * Get page information from URL
 * 
 * @param string $url Page URL
 * @return array|false Page data or false if not found
 */
function ai_ready_get_page_info_from_url($url) {
    $post_id = url_to_postid($url);
    
    if ($post_id) {
        $post = get_post($post_id);
        if ($post && $post->post_status === 'publish') {
            return [
                'id' => $post->ID,
                'title' => $post->post_title,
                'url' => $url,
                'type' => $post->post_type,
                'slug' => $post->post_name,
                'content_length' => strlen(wp_strip_all_tags($post->post_content))
            ];
        }
    }
    
    $path = wp_parse_url($url, PHP_URL_PATH);
    $slug = trim($path, '/');
    
    return [
        'id' => 0,
        'title' => ucwords(str_replace(['-', '_'], ' ', basename($slug))),
        'url' => $url,
        'type' => 'unknown',
        'slug' => $slug,
        'content_length' => 0
    ];
}

/**
 * Force refresh of llms.txt sections by clearing cached options
 */
function ai_ready_force_refresh_sections() {
    delete_option('ai_ready_llms_sections');
}

/**
 * Categorize pages by type/purpose for llms.txt sections
 * 
 * @param array $pages Array of page data from sitemap
 * @return array Categorized pages
 */
function ai_ready_categorize_sitemap_pages($pages) {
    $categories = [
        'docs' => [],
        'content' => [],
        'resources' => [],
        'optional' => []
    ];
    
    foreach ($pages as $page) {
        $slug = strtolower($page['slug']);
        $title = strtolower($page['title']);
        $category = ai_ready_determine_page_category($slug, $title, $page);
        
        if (!$category) {
            $category = 'content';
        }
        
        if ($category && isset($categories[$category])) {
            $categories[$category][] = $page;
        } else {
            $categories['content'][] = $page;
        }
    }
    
    return $categories;
}

/**
 * Determine which category a page belongs to
 * 
 * @param string $slug Page slug
 * @param string $title Page title (lowercase)
 * @param array $page_data Full page data
 * @return string Category name (always returns a category)
 */
function ai_ready_determine_page_category($slug, $title, $page_data) {
    $docs_keywords = [
        'about', 'mission', 'vision', 'who-we-are', 'our-story', 'company',
        'services', 'products', 'offerings', 'solutions', 'what-we-do',
        'contact', 'contact-us', 'get-in-touch', 'reach-us',
        'faq', 'help', 'support', 'frequently-asked-questions'
    ];
    
    foreach ($docs_keywords as $keyword) {
        if (strpos($slug, $keyword) !== false || strpos($title, $keyword) !== false) {
            return 'docs';
        }
    }
    
    $resource_keywords = [
        'resources', 'downloads', 'documentation', 'docs', 'guides',
        'tutorials', 'library', 'portfolio', 'work', 'projects',
        'case-studies', 'examples', 'showcase'
    ];
    
    foreach ($resource_keywords as $keyword) {
        if (strpos($slug, $keyword) !== false || strpos($title, $keyword) !== false) {
            return 'resources';
        }
    }
    
    if ($page_data['type'] === 'post' || $page_data['content_length'] > 1000) {
        return 'content';
    }
    
    if ($page_data['type'] === 'page') {
        return 'docs';
    }
    
    return 'optional';
}

/**
 * Check if llms.txt is enabled
 * 
 * @return bool True if llms.txt is enabled, false otherwise
 */
function ai_ready_is_llms_enabled() {
    if (get_option('ai_ready_llms_disabled', false)) {
        return false;
    }
    
    $has_title = !empty(get_option('ai_ready_site_title', ''));
    $has_summary = !empty(get_option('ai_ready_site_summary', ''));
    $has_sections = !empty(get_option('ai_ready_llms_sections', []));
    $has_custom = !empty(get_option('ai_ready_custom_content', ''));
    
    return true;
}