ImageProxy/ImageProxy.php

<?php
/**
 * 图片代理与缓存系统
 * 
 * 提供安全的图片代理和缓存服务，支持多种图片格式，具有完善的错误处理和日志记录功能
 */
class ImageProxy {
    // 支持的图片类型
    private const SUPPORTED_TYPES = [
        'gif' => 'image/gif',
        'jpeg' => 'image/jpeg',
        'jpg' => 'image/jpeg',
        'jpe' => 'image/jpeg',
        'png' => 'image/png',
        'webp' => 'image/webp',
        'bmp' => 'image/bmp',
        'svg' => 'image/svg+xml'
    ];

    // 默认配置
    private const DEFAULT_CONFIG = [
        'cache_dir' => 'cache',
        'timeout' => 30,
        'connect_timeout' => 15,
        'max_redirects' => 5,
        'user_agent' => 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36',
        'allowed_domains' => [] // 添加白名单域名配置
    ];

    private $config;
    private $logger;

    /**
     * 构造函数
     * 
     * @param array $config 配置参数
     */
    public function __construct(array $config = []) {
        $this->config = array_merge(self::DEFAULT_CONFIG, $config);
        $this->initCacheDir();
        $this->initLogger();
    }

    /**
     * 初始化缓存目录
     */
    private function initCacheDir() {
        if (!file_exists($this->config['cache_dir'])) {
            mkdir($this->config['cache_dir'], 0755, true);
        }
    }

    /**
     * 初始化日志记录器
     */
    private function initLogger() {
        $logDir = $this->config['cache_dir'] . '/logs';
        if (!file_exists($logDir)) {
            mkdir($logDir, 0777, true);
        }
        $this->logger = new Logger($logDir);
    }

    /**
     * 处理图片请求
     * 
     * @param string $url 图片URL
     * @return void
     */
    public function processRequest(string $url) {
        try {
            if (!$this->validateUrl($url)) {
                throw new Exception('Invalid URL format');
            }

            $cachePath = $this->getCachePath($url);
            
            if ($this->isCached($cachePath)) {
                $this->serveFromCache($cachePath);
                return;
            }

            $imageData = $this->fetchRemoteImage($url);
            
            $actualOriginalMimeType = null;
            if (!$this->validateImageData($imageData, $actualOriginalMimeType) || $actualOriginalMimeType === null) {
                $extensionBasedType = $this->getImageType($url); // For logging
                $this->logger->error("Validation failed for image data from URL: $url. Detected extension type: $extensionBasedType");
                throw new Exception('Invalid or unsupported image data from URL');
            }

            $dataToProcess = $imageData;
            $mimeTypeToUse = $actualOriginalMimeType;

            // Attempt conversion to WebP if applicable
            $convertedWebPData = $this->convertToWebP($dataToProcess, $actualOriginalMimeType);
            if ($convertedWebPData) {
                $dataToProcess = $convertedWebPData;
                $mimeTypeToUse = 'image/webp';
                $this->logger->info("Successfully converted to WebP: $url");
            }
            
            $this->saveToCache($cachePath, $dataToProcess, $mimeTypeToUse);
            $this->outputImage($dataToProcess, $mimeTypeToUse);

        } catch (Exception $e) {
            $this->logger->error('Image processing error for URL ' . $url . ': ' . $e->getMessage() . ' Trace: ' . $e->getTraceAsString());
            $this->outputError();
        }
    }

    /**
     * 验证URL格式
     * 
     * @param string $url URL地址
     * @return bool
     */
    private function validateUrl(string $url): bool {
        if (!filter_var($url, FILTER_VALIDATE_URL) || !preg_match('/^https?:\/\//i', $url)) {
            return false;
        }

        // 检查域名是否在白名单中
        $parsedUrl = parse_url($url);
        $host = $parsedUrl['host'] ?? '';
        
        // 如果白名单为空，则允许所有域名
        if (empty($this->config['allowed_domains'])) {
            return true;
        }

        // 检查域名是否在白名单中
        foreach ($this->config['allowed_domains'] as $allowedDomain) {
            if (strcasecmp($host, $allowedDomain) === 0 || 
                preg_match('/\.' . preg_quote($allowedDomain, '/') . '$/', $host)) {
                return true;
            }
        }

        $this->logger->error("Domain not in whitelist: $host");
        return false;
    }

    /**
     * 获取缓存路径
     * 
     * @param string $url 图片URL
     * @return string
     */
    private function getCachePath(string $url): string {
        $parsedUrl = parse_url($url);
        $domain = $parsedUrl['host'] ?? 'unknown_host'; // Add fallback for host
        $path = $parsedUrl['path'] ?? '/';
        
        $cacheSubDir = $this->config['cache_dir'] . '/' . preg_replace('/[^a-zA-Z0-9_\-\.]/', '_', $domain); // Sanitize domain for directory name
        if (!file_exists($cacheSubDir)) {
            mkdir($cacheSubDir, 0755, true);
        }
        
        $extension = pathinfo($path, PATHINFO_EXTENSION);
        if (empty($extension) && isset($parsedUrl['query'])) { // Try to get extension from query if path has none
            parse_str($parsedUrl['query'], $queryParams);
            foreach ($queryParams as $val) {
                $queryExt = pathinfo($val, PATHINFO_EXTENSION);
                if (!empty($queryExt) && isset(self::SUPPORTED_TYPES[strtolower($queryExt)])) {
                    $extension = $queryExt;
                    break;
                }
            }
        }
        $fileName = md5($url) . '.' . (!empty($extension) ? $extension : 'img'); // Fallback extension
        return $cacheSubDir . '/' . $fileName;
    }

    /**
     * 检查图片是否已缓存
     * 
     * @param string $cachePath 缓存路径
     * @return bool
     */
    private function isCached(string $cachePath): bool {
        $metaPath = $cachePath . '.meta';
        return file_exists($cachePath) &&
               file_exists($metaPath) && // Also check for meta file
               (time() - filemtime($cachePath)) < 86400; // 缓存24小时
    }

    /**
     * 从缓存输出图片
     * 
     * @param string $cachePath 缓存路径
     */
    private function serveFromCache(string $cachePath) {
        $metaPath = $cachePath . '.meta';
        $type = null;

        if (file_exists($metaPath)) {
            $type = trim(file_get_contents($metaPath));
        } else {
            $this->logger->error("Cache meta file missing for: $cachePath. Attempting to determine type from data.");
            if (file_exists($cachePath)) {
                $imageData = file_get_contents($cachePath);
                if ($imageData !== false && $this->validateImageData($imageData, $type) && $type !== null) {
                     file_put_contents($metaPath, $type); // Recreate meta
                     $this->logger->info("Recreated meta file for: $cachePath with type: $type");
                } else {
                    $this->logger->error("Could not determine type for cached file (meta missing): $cachePath");
                    $type = $this->getImageType($cachePath); // Fallback to extension of cache path itself
                }
            } else {
                $this->logger->error("Cache file missing during serveFromCache (isCached check failed?): $cachePath");
                $this->outputError();
                return;
            }
        }
        
        if (empty($type)) { 
            $this->logger->error("Unknown content type for cache file: $cachePath");
            $type = 'application/octet-stream'; // Generic fallback if type determination failed
        }

        header("Content-type: " . $type);
        header("Cache-Control: public, max-age=86400");
        readfile($cachePath);
    }

    /**
     * 获取远程图片
     * 
     * @param string $url 图片URL
     * @return string
     */
    private function fetchRemoteImage(string $url): string {
        $ch = curl_init($url);
        curl_setopt_array($ch, [
            CURLOPT_RETURNTRANSFER => true,
            CURLOPT_FOLLOWLOCATION => true,
            CURLOPT_MAXREDIRS => $this->config['max_redirects'],
            CURLOPT_CONNECTTIMEOUT => $this->config['connect_timeout'],
            CURLOPT_TIMEOUT => $this->config['timeout'],
            CURLOPT_SSL_VERIFYPEER => true, // Consider making this configurable for dev environments
            CURLOPT_SSL_VERIFYHOST => 2,
            CURLOPT_USERAGENT => $this->config['user_agent'],
            CURLOPT_REFERER => parse_url($url, PHP_URL_SCHEME) . '://' . (parse_url($url, PHP_URL_HOST) ?? ''),
            CURLOPT_HTTPHEADER => ['Accept: image/*,application/octet-stream'] // Accept more types
        ]);

        $data = curl_exec($ch);
        $httpCode = curl_getinfo($ch, CURLINFO_HTTP_CODE);
        $curlError = curl_error($ch);
        curl_close($ch);

        if ($curlError) {
            throw new Exception('cURL error fetching image: ' . $curlError . ' from URL: ' . $url);
        }
        if ($httpCode !== 200 || empty($data)) {
            throw new Exception('Failed to fetch image: HTTP ' . $httpCode . ' from URL: ' . $url);
        }

        return $data;
    }

    /**
     * 验证图片数据
     * 
     * @param string $data 图片数据
     * @param string|null &$outMimeType 输出参数，检测到的MIME类型
     * @return bool
     */
    private function validateImageData(string $data, ?string &$outMimeType = null): bool {
        if (empty($data)) {
            $outMimeType = null;
            return false;
        }

        // Order of checks can matter. WebP is quite specific.
        if (strlen($data) >= 12 && substr($data, 0, 4) === 'RIFF' && substr($data, 8, 4) === 'WEBP') {
            $outMimeType = 'image/webp';
            return true;
        }
        
        $signatures = [
            'image/jpeg' => "\xFF\xD8\xFF",
            'image/png'  => "\x89PNG\r\n\x1a\n",
            'image/gif'  => "GIF", // Covers GIF87a and GIF89a
            'image/bmp'  => "BM",
        ];

        foreach ($signatures as $mime => $sig) {
            if (substr($data, 0, strlen($sig)) === $sig) {
                $outMimeType = $mime;
                return true;
            }
        }
        
        // SVG check: more complex as it's XML-based.
        $trimmedDataStart = substr($data, 0, 256); // Check a small portion from the start for performance
        if (stripos($trimmedDataStart, '<?xml') === 0 || stripos($trimmedDataStart, '<svg') === 0) {
            if (stripos($data, '<svg') !== false && stripos($data, '</svg>') !== false) { // Basic check for opening and closing svg tag
                 $outMimeType = 'image/svg+xml';
                 return true;
            }
        }
        
        $outMimeType = null;
        return false;
    }

    /**
     * 转换图片为WebP格式
     * 
     * @param string $imageData 原始图片数据
     * @param string $originalMimeType 原始MIME类型
     * @return string|null WebP图片数据或null（如果转换失败或不支持）
     */
    private function convertToWebP(string $imageData, string $originalMimeType): ?string {
        if (!function_exists('imagewebp') || !function_exists('imagecreatefromstring')) {
            $this->logger->info('GD WebP support (imagewebp or imagecreatefromstring) not available.');
            return null; 
        }

        // Do not convert these types or if original is already WebP
        if (in_array($originalMimeType, ['image/webp', 'image/gif', 'image/svg+xml'])) {
            return null;
        }

        $sourceImage = @imagecreatefromstring($imageData);
        if (!$sourceImage) {
            $this->logger->error('imagecreatefromstring failed for ' . $originalMimeType . ' while attempting WebP conversion.');
            return null;
        }
        
        // Preserve alpha transparency for PNGs
        if ($originalMimeType === 'image/png') {
            imagealphablending($sourceImage, false);
            imagesavealpha($sourceImage, true);
        }

        ob_start();
        $success = imagewebp($sourceImage, null, 80); // Quality 80
        $webpData = ob_get_clean();
        imagedestroy($sourceImage);

        if (!$success || empty($webpData)) {
            $this->logger->error('imagewebp conversion failed for ' . $originalMimeType);
            return null;
        }
        
        // Minimal validation of the converted WebP data
        $validatedConvertedMimeType = null;
        if ($this->validateImageData($webpData, $validatedConvertedMimeType) && $validatedConvertedMimeType === 'image/webp') {
            return $webpData;
        } else {
            $this->logger->error('Converted WebP data failed validation. Original type: ' . $originalMimeType);
            return null;
        }
    }

    /**
     * 保存到缓存
     * 
     * @param string $cachePath 缓存路径
     * @param string $data 图片数据
     * @param string $actualMimeType 图片的实际MIME类型
     */
    private function saveToCache(string $cachePath, string $data, string $actualMimeType) {
        $parentDir = dirname($cachePath);
        if (!file_exists($parentDir)) {
            // This should ideally be handled by getCachePath, but as a safeguard:
            if (!mkdir($parentDir, 0755, true) && !is_dir($parentDir)) {
                 $this->logger->error("Failed to create cache subdirectory: $parentDir");
                 throw new Exception("Failed to create cache subdirectory: $parentDir");
            }
        }

        if (file_put_contents($cachePath, $data) === false) {
            $this->logger->error("Failed to write cache file: $cachePath");
            throw new Exception("Failed to write cache file: $cachePath");
        }
        if (file_put_contents($cachePath . '.meta', $actualMimeType) === false) {
            $this->logger->error("Failed to write meta file for: $cachePath . Mime: $actualMimeType");
            // If meta fails, cache is inconsistent. Clean up.
            unlink($cachePath); 
            throw new Exception("Failed to write cache meta file: $cachePath.meta");
        }
        $this->logger->info("Saved to cache: $cachePath, Type: $actualMimeType");
    }

    /**
     * 输出图片
     * 
     * @param string $data 图片数据
     * @param string $type 图片类型
     */
    private function outputImage(string $data, string $type) {
        header("Content-type: " . $type);
        header("Cache-Control: public, max-age=86400");
        echo $data;
    }

    /**
     * 获取图片类型 (基于文件扩展名)
     * 
     * @param string $urlOrPath 图片URL或路径
     * @return string
     */
    private function getImageType(string $urlOrPath): string {
        $ext = strtolower(pathinfo($urlOrPath, PATHINFO_EXTENSION));
        return self::SUPPORTED_TYPES[$ext] ?? 'application/octet-stream'; // Fallback type
    }

    /**
     * 输出错误信息
     */
    private function outputError() {
        // Check if headers already sent to avoid warning
        if (!headers_sent()) {
            header("HTTP/1.1 500 Internal Server Error");
            header("Content-type: image/jpeg"); // Keep consistent error image type
        }
        // Output a 1x1 transparent GIF as error image
        echo base64_decode('R0lGODlhAQABAIAAAAAAAP///yH5BAEAAAAALAAAAAABAAEAAAIBRAA7');
    }
}

/**
 * 简单的日志记录类
 */
class Logger {
    private $logDir;
    private const MAX_LOG_SIZE = 10 * 1024 * 1024; // 10MB
    private const MAX_LOG_AGE = 30 * 24 * 60 * 60; // 30天
    private const MAX_LOGS = 10; // 最多保留10个日志文件

    public function __construct(string $logDir) {
        $this->logDir = $logDir;

        if (!file_exists($this->logDir)) {
            mkdir($this->logDir, 0777, true);
        }
        
        // 清理旧日志
        $this->cleanupOldLogs();
    }

    public function error(string $message) {
        $this->log('ERROR', $message);
    }

    public function info(string $message) {
        $this->log('INFO', $message);
    }

    private function cleanupOldLogs() {
        try {
            $files = glob($this->logDir . '/*.log');
            if (!$files) {
                return;
            }

            // 按修改时间排序
            usort($files, function($a, $b) {
                return filemtime($b) - filemtime($a);
            });

            // 删除超过保留时间的日志
            foreach ($files as $file) {
                if (time() - filemtime($file) > self::MAX_LOG_AGE) {
                    unlink($file);
                }
            }

            // 如果日志文件数量超过限制，删除最旧的
            $files = glob($this->logDir . '/*.log');
            if (count($files) > self::MAX_LOGS) {
                usort($files, function($a, $b) {
                    return filemtime($a) - filemtime($b);
                });
                $filesToDelete = array_slice($files, 0, count($files) - self::MAX_LOGS);
                foreach ($filesToDelete as $file) {
                    unlink($file);
                }
            }
        } catch (Exception $e) {
            error_log("Logger cleanup error: " . $e->getMessage());
        }
    }

    private function rotateLogIfNeeded(string $logFile) {
        if (!file_exists($logFile)) {
            return;
        }

        if (filesize($logFile) >= self::MAX_LOG_SIZE) {
            $timestamp = date('Y-m-d_H-i-s');
            $newName = $logFile . '.' . $timestamp;
            rename($logFile, $newName);
        }
    }

    private function log(string $level, string $message) {
        try {
            $date = date('Y-m-d');
            $time = date('Y-m-d H:i:s');
            $logFile = $this->logDir . "/{$date}.log";
            
            // 检查并轮转日志
            $this->rotateLogIfNeeded($logFile);
            
            $logMessage = "[{$time}] [{$level}] {$message}\n";
            
            // 确保日志目录可写
            if (!is_writable($this->logDir) && !mkdir($this->logDir, 0777, true) && !is_dir($this->logDir)) {
                error_log("Logger Error: Directory {$this->logDir} is not writable. Message: {$logMessage}");
                return;
            }

            // 写入日志
            if (file_put_contents($logFile, $logMessage, FILE_APPEND) === false) {
                error_log("Logger Error: Failed to write to log file {$logFile}. Message: {$logMessage}");
            }
        } catch (Exception $e) {
            error_log("Logger Exception: " . $e->getMessage() . " Original log message: {$message}");
        }
    }
}