PHP导出含图片Word文档的MHT原理与实现

/**
 * 根据HTML代码获取word文档内容
 * 创建一个本质为mht的文档，该函数会分析文件内容并从远程下载页面中的图片资源
 *
 * @param string $content HTML内容
 * @param string $absolutePath 网页的绝对路径。如果HTML内容里的图片路径为相对路径，那么就需要填写这个参数，来让该函数自动填补成绝对路径。这个参数最后需要以/结束
 * @param bool $isEraseLink 是否去掉HTML内容中的链接
 */
function getWordDocument( $content , $absolutePath = "" , $isEraseLink = true ) {
    $mht = new MhtFileMaker();
    
    // 去掉链接 (可选)
    if ($isEraseLink)
        $content = preg_replace('/<a\s*.*?\s*>(\s*.*?\s*)<\/a>/i' , '$1' , $content);

    $images = array();
    $files = array();
    $matches = array();
    
    // 提取img标签中的src属性 (要求src值必须被引号包围)
    if ( preg_match_all('/<img[.\n]*?src\s*?=\s*[\"\']([^"]+)[\"\'](.*?)\/>/i', $content, $matches ) ) {
        $arrPath = $matches[1];
        for ( $i=0; $i<count($arrPath); $i++ ) {
            $path = $arrPath[$i];
            $imgPath = trim( $path );
            if ( $imgPath != "" ) {
                $files[] = $imgPath;
                
                // 判断是否为绝对路径
                if( substr($imgPath,0,7) == 'http://') {
                    // 已是绝对链接，不加前缀
                } else {
                    // 相对路径，拼接绝对路径
                    $imgPath = $absolutePath.$imgPath;
                }
                $images[] = $imgPath;
            }
        }
    }
    
    // 添加主HTML内容
    $mht->AddContents("tmp.html", $mht->GetMimeType("tmp.html"), $content);
    
    // 遍历并添加图片资源
    for ( $i=0; $i<count($images); $i++ ) {
        $image = $images[$i];
        // 尝试打开图片流
        if ( @fopen($image , 'r') ) {
            $imgcontent = @file_get_contents( $image );
            if ( $imgcontent ) {
                $mht->AddContents($files[$i], $mht->GetMimeType($image), $imgcontent);
            }
        } else {
            // 记录错误或忽略
            echo "file:".$image." not exist!<br />";
        }
    }
    
    return $mht->GetFile();
}

?php
class MhtFileMaker{
    var $config = array();
    var $headers = array();
    var $headers_exists = array();
    var $files = array();
    var $boundary;
    var $dir_base;
    var $page_first;

    function MhtFile($config = array()){
        // 初始化配置
    }

    function SetHeader($header){
        $this->headers[] = $header;
        $key = strtolower(substr($header, 0, strpos($header, ':')));
        $this->headers_exists[$key] = TRUE;
    }

    function SetFrom($from){
        $this->SetHeader("From: $from");
    }

    function SetSubject($subject){
        $this->SetHeader("Subject: $subject");
    }

    function SetDate($date = NULL, $istimestamp = FALSE){
        if ($date == NULL) {
            $date = time();
        }
        if ($istimestamp == TRUE) {
            $date = date('D, d M Y H:i:s O', $date);
        }
        $this->SetHeader("Date: $date");
    }

    function SetBoundary($boundary = NULL){
        if ($boundary == NULL) {
            $this->boundary = '--' . strtoupper(md5(mt_rand())) . '_MULTIPART_MIXED';
        } else {
            $this->boundary = $boundary;
        }
    }

    function SetBaseDir($dir){
        $this->dir_base = str_replace("\\", "/", realpath($dir));
    }

    function SetFirstPage($filename){
        $this->page_first = str_replace("\\", "/", realpath("{$this->dir_base}/$filename"));
    }

    // ... (此处省略内部私有辅助方法如 AutoAddFiles, AddDir 等，它们主要用于文件系统操作) ...

    /**
     * 添加文件内容到集合中
     */
    function AddContents($filepath, $mimetype, $filecont, $encoding = NULL){
        if ($encoding == NULL) {
            // 默认使用 Base64 编码
            $filecont = chunk_split(base64_encode($filecont), 76);
            $encoding = 'base64';
        }
        $this->files[] = array(
            'filepath' => $filepath,
            'mimetype' => $mimetype,
            'filecont' => $filecont,
            'encoding' => $encoding
        );
    }

    /**
     * 检查必要的 Headers 和 Boundary
     */
    function CheckHeaders(){
        if (!array_key_exists('date', $this->headers_exists)) {
            $this->SetDate(NULL, TRUE);
        }
        if ($this->boundary == NULL) {
            $this->SetBoundary();
        }
    }

    function GetMimeType($filename){
        $pathinfo = pathinfo($filename);
        switch ($pathinfo['extension']) {
            case 'htm': case 'html': return 'text/html'; break;
            case 'txt': case 'cgi': case 'php': return 'text/plain'; break;
            case 'css': return 'text/css'; break;
            case 'jpg': case 'jpeg': case 'jpe': return 'image/jpeg'; break;
            case 'gif': return 'image/gif'; break;
            case 'png': return 'image/png'; break;
            default: return 'application/octet-stream'; break;
        }
    }

    /**
     * 生成最终的 MHT 字符串
     */
    function GetFile(){
        $this->CheckHeaders();
        
        $contents = implode("\r\n", $this->headers);
        $contents .= "\r\n";
        $contents .= "MIME-Version: 1.0\r\n";
        $contents .= "Content-Type: multipart/related;\r\n";
        $contents .= "\tboundary=\"{$this->boundary}\";\r\n";
        $contents .= "\ttype=\"" . $this->files[0]['mimetype'] . "\"\r\n";
        $contents .= "X-MimeOLE: Produced By Mht File Maker v1.0 beta\r\n";
        $contents .= "\r\n";
        $contents .= "This is a multi-part message in MIME format.\r\n";
        $contents .= "\r\n";
        
        foreach ($this->files as $file) {
            $contents .= "--{$this->boundary}\r\n";
            $contents .= "Content-Type: $file[mimetype]\r\n";
            $contents .= "Content-Transfer-Encoding: $file[encoding]\r\n";
            $contents .= "Content-Location: $file[filepath]\r\n";
            $contents .= "\r\n";
            $contents .= $file['filecont'];
            $contents .= "\r\n";
        }
        
        $contents .= "--{$this->boundary}--\r\n";
        return $contents;
    }
}
?>

// $content 是你的 HTML 源代码
// 第二个参数是图片所在的基础URL，确保相对路径能转为绝对路径
$fileContent = getWordDocument($content, "https://example.com/images/");

// 保存为文件
$fp = fopen("test.doc", 'w');
fwrite($fp, $fileContent);
fclose($fp);