Extracto Web & Image Analyzer

<?php
/**
 * PRODUCTION-GRADE WEB CRAWLER & EMAIL EXTRACTOR WITH MYSQL INTEGRATION & SERVER SIDE OCR
 * Language: PHP 8.0+ (cURL, DOMDocument, PDO, Tesseract-OCR)
 * Composer installation: composer require thiagoalessio/tesseract-ocr
 */

header('Content-Type: text/html; charset=utf-8');

// --- DATABASE CONFIGURATION ---
define('DB_HOST', 'localhost');
define('DB_NAME', 'email_crawler_db');
define('DB_USER', 'root');
define('DB_PASS', '');

try {
    $db = new PDO("mysql:host=" . DB_HOST, DB_USER, DB_PASS, [
        PDO::ATTR_ERRMODE => PDO::ERRMODE_EXCEPTION,
        PDO::ATTR_DEFAULT_FETCH_MODE => PDO::FETCH_ASSOC
    ]);
    $db->exec("CREATE DATABASE IF NOT EXISTS " . DB_NAME);
    $db->exec("USE " . DB_NAME);
    
    $db->exec("CREATE TABLE IF NOT EXISTS crawl_jobs (
        id INT AUTO_INCREMENT PRIMARY KEY,
        start_url VARCHAR(255) NOT NULL,
        max_depth INT DEFAULT 2,
        pages_limit INT DEFAULT 25,
        created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
    ) ENGINE=InnoDB;");
    
    $db->exec("CREATE TABLE IF NOT EXISTS extracted_emails (
        id INT AUTO_INCREMENT PRIMARY KEY,
        job_id INT,
        email VARCHAR(150) NOT NULL,
        source_url VARCHAR(255) NOT NULL,
        extraction_type VARCHAR(50) DEFAULT 'DOM Text',
        discovered_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
        FOREIGN KEY (job_id) REFERENCES crawl_jobs(id) ON DELETE CASCADE,
        UNIQUE KEY unique_email_url (email, source_url)
    ) ENGINE=InnoDB;");
} catch (PDOException $e) {
    die("Database Connection Failed: " . htmlspecialchars($e->getMessage()));
}

// --- CORE CRAWLING & OCR LOGIC ---
use thiagoalessio\TesseractOCR\TesseractOCR;

class ServerEmailCrawler {
    private $startUrl;
    private $maxDepth;
    private $maxPages;
    private $jobId;
    private $pdo;
    private $enableOcr = true;
    private $visited = [];
    private $pagesProcessed = 0;

    public function __construct($pdo, $startUrl, $maxDepth = 2, $maxPages = 20, $enableOcr = true) {
        $this->pdo = $pdo;
        $this->startUrl = rtrim(filter_var($startUrl, FILTER_SANITIZE_URL), '/');
        $this->maxDepth = intval($maxDepth);
        $this->maxPages = intval($maxPages);
        $this->enableOcr = $enableOcr;

        $stmt = $this->pdo->prepare("INSERT INTO crawl_jobs (start_url, max_depth, pages_limit) VALUES (?, ?, ?)");
        $stmt->execute([$this->startUrl, $this->maxDepth, $this->maxPages]);
        $this->jobId = $this->pdo->lastInsertId();
    }

    public function start() {
        echo "Starting crawl on " . htmlspecialchars($this->startUrl) . "\n";
        $this->crawl($this->startUrl, 1);
    }

    private function crawl($url, $depth) {
        if ($depth > $this->maxDepth || $this->pagesProcessed >= $this->maxPages || isset($this->visited[$url])) {
            return;
        }
        $this->visited[$url] = true;
        $this->pagesProcessed++;

        $html = $this->fetchWithCurl($url);
        if (!$html) return;

        // 1. Text Extraction
        $emails = $this->extractEmails($html);
        foreach ($emails as $email) {
            $this->saveEmail($email, $url, 'DOM HTML Text');
        }

        // 2. OCR Image Extraction
        if ($this->enableOcr) {
            $images = $this->extractImages($html, $url);
            foreach ($images as $imgUrl) {
                $imgData = $this->fetchWithCurl($imgUrl);
                if ($imgData) {
                    $tmpFile = sys_get_temp_dir() . DIRECTORY_SEPARATOR . 'ocr_tmp_' . md5($imgUrl);
                    file_put_contents($tmpFile, $imgData);
                    try {
                        // Scan downloaded image using local tesseract binaries
                        $ocrText = (new TesseractOCR($tmpFile))->run();
                        $ocrEmails = $this->extractEmails($ocrText);
                        foreach ($ocrEmails as $ocrEmail) {
                            $this->saveEmail($ocrEmail, $url, 'OCR Image Extraction (' . basename($imgUrl) . ')');
                        }
                    } catch (\Exception $e) {
                        // Failed to OCR (e.g. invalid format or binaries missing)
                    }
                    @unlink($tmpFile);
                }
            }
        }

        // Recursion
        if ($depth < $this->maxDepth) {
            $links = $this->extractLinks($html, $url);
            foreach ($links as $link) {
                $this->crawl($link, $depth + 1);
            }
        }
    }

    private function fetchWithCurl($url) {
        $ch = curl_init();
        curl_setopt($ch, CURLOPT_URL, $url);
        curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
        curl_setopt($ch, CURLOPT_FOLLOWLOCATION, true);
        curl_setopt($ch, CURLOPT_SSL_VERIFYPEER, false);
        curl_setopt($ch, CURLOPT_TIMEOUT, 10);
        $res = curl_exec($ch);
        curl_close($ch);
        return $res;
    }

    private function extractEmails($text) {
        preg_match_all('/[a-z0-9._%+-]+@[a-z0-9.-]+\.[a-z]{2,6}/i', $text, $matches);
        return array_unique($matches[0] ?? []);
    }

    private function extractImages($html, $baseUrl) {
        $images = [];
        $dom = new DOMDocument();
        @$dom->loadHTML($html);
        foreach ($dom->getElementsByTagName('img') as $img) {
            $src = $img->getAttribute('src');
            if ($src) {
                if (strpos($src, 'http') !== 0) {
                    $src = rtrim($baseUrl, '/') . '/' . ltrim($src, '/');
                }
                $images[] = $src;
            }
        }
        return array_slice(array_unique($images), 0, 10); // Limit to 10 images per page
    }

    private function extractLinks($html, $baseUrl) {
        $links = [];
        $dom = new DOMDocument();
        @$dom->loadHTML($html);
        foreach ($dom->getElementsByTagName('a') as $a) {
            $href = $a->getAttribute('href');
            if ($href && strpos($href, '#') !== 0 && strpos($href, 'mailto:') !== 0) {
                if (strpos($href, 'http') !== 0) {
                    $href = rtrim($baseUrl, '/') . '/' . ltrim($href, '/');
                }
                $links[] = $href;
            }
        }
        return array_slice(array_unique($links), 0, 15);
    }

    private function saveEmail($email, $url, $type) {
        try {
            $stmt = $this->pdo->prepare("INSERT IGNORE INTO extracted_emails (job_id, email, source_url, extraction_type) VALUES (?, ?, ?, ?)");
            $stmt->execute([$this->jobId, $email, $url, $type]);
        } catch (\PDOException $e) {}
    }
}
?>