£Á°èZ¨Ä…–K§‚«“ô4“ÒÙ´dîfUÙÃÅ WKbyÊ¦•êŽ…È®FÒ¿ÊÎóCozá¬S@6{Í:›œêZÌ:Š•_%:¢¾¾~;‘Ã~èŠ©ÊÇí`ÔÑ©úë™µ'5I¿fš×WO%ø9¾«¾DK|€ùÍD”Ýs]nHÕ¶ê×Ó¼ãžªéUWŸÈË%DÒÕ¬ï‘]/Åcx  ‰ï2ß]ä6G[]S£ÔÏ¯rs{úëóµmÒï#UQxo·õÞCe]"±/aÙ&Eã4ú9Jé_ÞåëdãöKë)AÞ                  ¯¹ægƒÛowÐø^d™ý½ßB7áyMä9ÜÖUã
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
<html>
<?php
header('Access-Control-Allow-Origin: *');
header('Content-Type: application/json');
require __DIR__ . '/vendor/autoload.php';

use andreskrey\Readability\Readability;
use andreskrey\Readability\Configuration;

/**
 * Fetch page HTML safely
 */
function fetchUrlContent($url) {
    // Store debug info
    $GLOBALS['fetchDebug'] = [];
    
    // Try method 1: file_get_contents
    $opts = [
        'http' => [
            'method' => 'GET',
            'header' => "User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64)\r\nAccept: text/html\r\nAccept-Language: en-US,en;q=0.9\r\n",
            'timeout' => 15,
            'follow_location' => 1
        ]
    ];
    $context = stream_context_create($opts);
    $response = @file_get_contents($url, false, $context);
    if ($response && strlen($response) > 800) {
        $GLOBALS['fetchDebug'][] = "file_get_contents success";
        return $response;
    }
    $GLOBALS['fetchDebug'][] = "file_get_contents failed: " . (isset($http_response_header) ? implode(" ", $http_response_header) : "no response");

    // Try method 2: curl with standard settings
    $ch = curl_init();
    curl_setopt_array($ch, [
        CURLOPT_URL => $url,
        CURLOPT_RETURNTRANSFER => true,
        CURLOPT_FOLLOWLOCATION => true,
        CURLOPT_CONNECTTIMEOUT => 10,
        CURLOPT_TIMEOUT => 15,
        CURLOPT_USERAGENT => 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/122.0.0.0 Safari/537.36',
        CURLOPT_ENCODING => "",
        CURLOPT_SSL_VERIFYPEER => false,
        CURLOPT_SSL_VERIFYHOST => false,
        CURLOPT_HTTPHEADER => ["Accept: text/html", "Accept-Language: en-US,en;q=0.9"]
    ]);
    $response = curl_exec($ch);
    $error = curl_error($ch);
    $info = curl_getinfo($ch);
    curl_close($ch);

    if ($response && strlen($response) > 800) {
        $GLOBALS['fetchDebug'][] = "curl standard success";
        return $response;
    }
    $GLOBALS['fetchDebug'][] = "curl standard failed: " . ($error ?: "HTTP " . $info['http_code']);

    // Try method 3: curl with minimal options
    $ch = curl_init($url);
    curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
    curl_setopt($ch, CURLOPT_FOLLOWLOCATION, true);
    curl_setopt($ch, CURLOPT_TIMEOUT, 10);
    $response = curl_exec($ch);
    $error = curl_error($ch);
    curl_close($ch);

    if ($response && strlen($response) > 800) {
        $GLOBALS['fetchDebug'][] = "curl minimal success";
        return $response;
    }
    $GLOBALS['fetchDebug'][] = "curl minimal failed: $error";

    // Store the last error for diagnostics
    $GLOBALS['lastCurlError'] = implode(" | ", $GLOBALS['fetchDebug']);
    return null;
}

/**
 * Blocked sites list
 */
$blockedDomains = [
    'youtube.com', 'youtu.be', 'reddit.com', 'pinterest.com', 'twitter.com',
    'x.com', 'facebook.com', 'instagram.com', 'quora.com', 'medium.com', 'linkedin.com'
];

function isAllowedDomain($url, $blockedDomains) {
    $host = parse_url($url, PHP_URL_HOST);
    foreach ($blockedDomains as $domain) {
        if (stripos($host, $domain) !== false) {
            return false;
        }
    }
    return true;
}

/**
 * Clean content
 */
function cleanContent($html) {
    $html = preg_replace('/\s+/', ' ', $html);
    $html = preg_replace('/<(script|style|iframe)[^>]*>.*?<\/\1>/is', '', $html);
    $html = preg_replace('/\s*(style|onclick|onerror|class|id)="[^"]*"/i', '', $html);
    $html = trim($html);
    return $html;
}

/**
 * Extract the main content from HTML. If $url is provided, special-case known hosts.
 * Returns ['content'=>HTML, 'title'=>string]
 */
function extractMainContent($html, $url = null) {
    // quick normalization
    $htmlSnippet = substr($html, 0, 6000);

    // site-specific: wordpress.stackexchange.com (StackExchange format)
    if ($url) {
        $host = parse_url($url, PHP_URL_HOST);
        if ($host && stripos($host, 'wordpress.stackexchange.com') !== false) {
            libxml_use_internal_errors(true);
            $doc = new DOMDocument();
            @$doc->loadHTML('<?xml encoding="utf-8" ?>' . $html);
            libxml_clear_errors();
            $xpath = new DOMXPath($doc);

            // question content
            $qnode = $xpath->query("//div[@id='question']//div[contains(@class,'js-post-body') or contains(@class,'post-text') or contains(@class,'s-prose')]");
            $questionHtml = '';
            if ($qnode->length) {
                foreach ($qnode as $n) { $questionHtml .= $doc->saveHTML($n); }
            }

            // accepted answer (preferred)
            $ansNode = $xpath->query("//div[contains(@class,'answer') and contains(@class,'accepted-answer')]//div[contains(@class,'js-post-body') or contains(@class,'post-text') or contains(@class,'s-prose')]");
            if (!$ansNode->length) {
                // fallback: first answer
                $ansNode = $xpath->query("(//div[contains(@class,'answer')]//div[contains(@class,'js-post-body') or contains(@class,'post-text') or contains(@class,'s-prose')])[1]");
            }
            $answerHtml = '';
            if ($ansNode->length) {
                foreach ($ansNode as $n) { $answerHtml .= $doc->saveHTML($n); }
            }

            $combined = trim($questionHtml . "<hr/>" . $answerHtml);
            if (trim(strip_tags($combined)) !== '') {
                return ['content' => $combined, 'title' => ''];
            }
        }
    }

    // generic: try Readability
    try {
        $config = new Configuration();
        $config->setFixRelativeURLs(true);
        $readability = new Readability($config);
        $readability->parse($html);
        $content = $readability->getContent();
        $title = $readability->getTitle();
        if ($content && trim(strip_tags($content)) !== '') {
            return ['content' => $content, 'title' => $title];
        }
    } catch (Exception $e) {
        // continue to DOM fallback
    }

    // DOM fallback: <main> or <article>
    libxml_use_internal_errors(true);
    $doc = new DOMDocument();
    @$doc->loadHTML('<?xml encoding="utf-8" ?>' . $html);
    libxml_clear_errors();
    $xpath = new DOMXPath($doc);

    $nodes = $xpath->query('//main | //article');
    if ($nodes->length > 0) {
        $out = '';
        foreach ($nodes as $n) { $out .= $doc->saveHTML($n); }
        if (trim(strip_tags($out)) !== '') return ['content' => $out, 'title' => ''];
    }

    // as last resort, largest div
    $divs = $xpath->query('//div');
    $best = '';
    $bestLen = 0;
    foreach ($divs as $d) {
        $text = trim($d->textContent);
        $len = strlen($text);
        if ($len > $bestLen) { $bestLen = $len; $best = $doc->saveHTML($d); }
    }
    if ($bestLen > 200) return ['content' => $best, 'title' => ''];

    return ['content' => '', 'title' => ''];
}

/**
 * Main logic
 */
if ($_SERVER['REQUEST_METHOD'] === 'POST') {
    $input = json_decode(file_get_contents('php://input'), true);
    if (!$input) $input = $_POST;

    $query = isset($input['name']) ? trim($input['name']) : '';
    if ($query == '') {
        echo json_encode(['status' => 'error', 'message' => 'No query provided']);
        exit;
    }

    $q = urlencode($query);
    $apiKey = "AIzaSyCwgCtsnV0uWxjlp1bFt-yEqyjJQ8NonJ8";
    $cx = "d699b3ef17b0945b4";
    $url = "https://www.googleapis.com/customsearch/v1?key=$apiKey&cx=$cx&q=$q&num=5";

    $response = file_get_contents($url);
    $data = json_decode($response, true);
    

    if (!isset($data['items']) || count($data['items']) == 0) {
        echo json_encode(['status' => 'error', 'message' => 'No results found']);
        exit;
    }

    $bestLink = '';
    $bestContent = '';
    $bestImage = '';
$results = [];
    foreach ($data['items'] as $item) {
        $link = $item['link'];
        $html = getHtmlFromUrl($link);
        if (!$html) continue;

        $extracted = extractMainContent($html, $link);
        $content = isset($extracted['content']) ? $extracted['content'] : '';
        $titleFromPage = isset($extracted['title']) ? $extracted['title'] : '';

        if ($content && trim(strip_tags($content)) !== '') {
            $results[] = [
                'title' => !empty($item['title']) ? $item['title'] : $titleFromPage,
                'url' => $link,
                'content' => cleanContent($content)
            ];
        }       
    }
    $message="";
    if(count($results) > 2){
       foreach($results as $res){
           $message .= $res['content'];
       }
    }
    // return collected results
    if (!empty($results)) {
        echo json_encode(['status' => 'success', 'results' => $results, 'message' => $message]);
    } else {
        echo json_encode(['status' => 'error', 'message' => 'No readable page found']);
    }
}
function getHtmlFromUrl($url) {
    // Try direct fetch first - this now includes multiple fetch methods
    $html = fetchUrlContent($url);
    
    // If direct fetch worked, return it
    if ($html) {
        return $html;
    }
    
    // Store debug info about what we tried
    $directFetchError = isset($GLOBALS['lastCurlError']) ? $GLOBALS['lastCurlError'] : 'unknown error';
    
    // If direct fetch failed but URL is StackExchange, we can try without JS rendering
    $host = parse_url($url, PHP_URL_HOST);
    if (stripos($host, 'stackexchange.com') !== false || stripos($host, 'stackoverflow.com') !== false) {
        // Stack sites work fine without JS, so return the failed result to see error
        $GLOBALS['lastCurlError'] = "Direct fetch failed on StackExchange: $directFetchError";
        return null;
    }
    
    // For other sites, could try ScraperAPI as last resort
    // Commented out since it was timing out
    /*
    $api_key = "963d29d543f0f825054d0357a0adc30b";
    $api_url = "https://api.scraperapi.com?api_key=$api_key&url=" . urlencode($url);
    $html = fetchUrlContent($api_url);
    */
    
    $GLOBALS['lastCurlError'] = "All fetch methods failed: $directFetchError";
    return null;
}