优化文件下载

时间:2014-04-07 18:43:26

标签: java http get

我正在尝试从服务器下载大约129个报告时优化我的代码。所有报告都有不同的URL,这是我开发的代码:

public static void getReport(String eqID, String reportCode, String reportName, String fileName) throws IOException{

        String url = "http://example.com/api?function=getData&eqId=" + eqID;
        URL obj = new URL(url);
        HttpURLConnection con = (HttpURLConnection) obj.openConnection();

        Authenticator.setDefault (new Authenticator() {
            protected PasswordAuthentication getPasswordAuthentication() {
                return new PasswordAuthentication ("username", "password".toCharArray());
            }
        });

        // optional default is GET
        con.setRequestMethod("GET");

        int responseCode = con.getResponseCode();

        if(responseCode == 200){
            System.out.println("Downloading: " + reportName);
            File file = new File("C:/Users/fileName);

            BufferedInputStream bis = new BufferedInputStream(con.getInputStream());
            BufferedOutputStream bos = new BufferedOutputStream(new FileOutputStream(file.getPath()));

            int i = 0;
            while ((i = bis.read()) != -1) {
                bos.write(i);
            }

            bos.flush();
            bis.close();
            bos.close();
        }
        else if(responseCode == 204){
            System.out.println("\nSending 'GET' request to: " + reportName);
            System.out.println("Response: Successful but report is empty. It will be skipped");
        }
    }

问题在于它花费了太多时间来处理这些下载。在所有129个报告中,它大约25MB,我有一个高速互联网连接。我通过java下载文件相当新,需要一些帮助。我总共称这种方法为129次。

如果您可以推荐优化方法或仅使用HTTP连接而不是单独打开129。

提前致谢!

2 个答案:

答案 0 :(得分:3)

瓶颈在这里:

int i = 0;
while ((i = bis.read()) != -1) {
    bos.write(i);
}

您正在逐字节读取,这可能会在较大的文件中花费大量时间。相反,通过块读取文件,通常是4KB或8KB:

int FILE_CHUNK_SIZE = 1024 * 4; //to make it easier to change to 8 KBs
byte[] chunk = new byte[FILE_CHUNK_SIZE];
int bytesRead = 0;
while ((bytesRead = input.read(chunk)) != -1) {
    bos.write(chunk, 0, bytesRead);
}

另一个替代方案是使用IOUtils#copy中的Apache Commons IO已经为您执行此操作:

IOUtils.copy(bis, bos);

答案 1 :(得分:0)

如果您使用的是Java 7,则非常简单。如果要编写文件,请执行以下操作:

final Path dstFile = Paths.get("C:/users/filename");

if (responseCode == 200) {
    try (
        final InputStream in = con.getInputStream();
    ) {
        Files.copy(in, dstFile, StandardOpenOption.CREATE_NEW);
    }
} // etc