为什么恢复下载会在 Java 应用程序中下载损坏的文件？答案

【问题标题】：Why Resume Downloading downloads corrupted file in Java Application?为什么恢复下载会在 Java 应用程序中下载损坏的文件？
【发布时间】：2018-12-31 11:50:42
【问题描述】：

在我的 Struts 基础 Java Web 应用程序中，我实现了带有暂停/恢复的下载功能。让我解释一下我的下载是如何工作的，实际上文件（例如图像文件，doc）将在服务器中转换为 pdf 然后它将开始下载，然后没有暂停，它已成功下载文件并且它没有损坏但是当我暂停并恢复时它成功下载文件，但它会损坏pdf。我想不出我错过了什么。我的代码如下

{           in = new FileInputStream(f);
            int DOWNLOAD_BUFFER_SIZE = Integer.parseInt(CacheManagement.getInstance().getSystemPropertyByAlias().get("DOWNLOAD_BUFFER_SIZE").getPropertyValue());

            response.setHeader("Accept-Ranges", "bytes");
            response.setContentType("application/x-download");
            response.setHeader("Content-Disposition", "attachment; filename=\"" + fileName + "\"");
            response.setBufferSize(DOWNLOAD_BUFFER_SIZE);

            // SimpleDateFormat formatter = new SimpleDateFormat("EEE, d MMM yyyy HH:mm:ss Z");
            // Date lastModifiedDate = formatter.parse(String.valueOf(f.lastModified()));
            Date lastModifiedDate = new Date(f.lastModified());
            response.setHeader("ETag", f.length() + "_" + lastModifiedDate.getTime());
            response.setDateHeader("Last-Modified", lastModifiedDate.getTime());

            String range = request.getHeader("Range");
            if (range != null && !range.isEmpty())
            {
                ArrayList<Range> ranges = getRangeList(range, f.length());
                Range r = ranges.get(0);
                response.setHeader("Content-Range", "bytes " + r.getStart() + "-" + r.getEnd() + "/" + r.getTotal());
                response.setHeader("Content-Length", String.valueOf(r.getLength()));
                response.setStatus(HttpServletResponse.SC_PARTIAL_CONTENT);
            }
            else
            {
                response.setHeader("Content-Length", String.valueOf(f.length()));
                response.setHeader("Content-Range", "bytes 0-" + (f.length() - 1) + "/" + f.length());
            }

            o = response.getOutputStream();
            buf = new BufferedInputStream(in);
            byte[] dataBytes = new byte[DOWNLOAD_BUFFER_SIZE];
            int readed = 0;
            while ((readed = buf.read(dataBytes)) != -1)
            {
                try
                {
                    o.write(dataBytes, 0, readed);
                }
                catch (Exception e)
                {
                    // System.out.println("-- Client Aborted Download --");
                    logger.error("-- Client Aborted Download --");
                    break;
                }
            }

【问题讨论】：

标签： java io struts2 inputstream download

【解决方案1】：

不要使用 buf = new BufferedInputStream(in); 来读取剩余的字节，你需要使用 RandomAccessFile 类和 seek() 使用剩余字节开始下载的方法。

    RandomAccessFile raFile = new RandomAccessFile(f, "r");
    o = response.getOutputStream();
    // buf = new BufferedInputStream(in);
    byte[] dataBytes = new byte[DOWNLOAD_BUFFER_SIZE];
    int readed;
    raFile.seek(r.getStart());
    while ((readed = raFile.read(dataBytes)) != -1)
    {
        try
        {
            o.write(dataBytes, 0, readed);
        }
        catch (Exception e)
        {
            logger.error("-- Client Aborted Download --");
            break;
        }
    }

【讨论】：