Servlets 无法下载小文件，但可以下载大文件

我正在使用下面的代码从S3下载文件。该代码适用于中到大的文件大小，但是，对于3kb的非常小的文件大小-TXT文件中的一行，下载失败

//控制器

def download() {
   Request request = Request.get(params.int("id"))
   response.setContentType("application/octet-stream")
   myService.downloadFileFromS3(request.origFileName, response)
}


void downloadFileFromS3(String fileName, HttpServletResponse response) {
    String fullFileNameWithExtension = fileName
    response.setHeader("Content-disposition", "attachment;filename=${fullFileNameWithExtension}")
    InputStream is = getS3Client().getObject(getBucketName(), fullFileNameWithExtension).getObjectContent()
    OutputStream outputStream = response.getOutputStream()
    byte[] buffer = new byte[1024];
    int length
    log.info("This is the length: " + length)
    while ((length = is.read(buffer)) != -1) {
        log.info("Came here with length: " + length)
        outputStream.write(buffer, 0, length)
    }
    is.close()
}

小文件的注销：

This is the length: 0
Came here with length: 15
GroovyPagesServlet:  "/WEB-INF/grails-app/views/request/download.gsp" not found

大型文件的日志输出：

This is the length: 0
Came here with length: 1024
Came here with length: 1024
Came here with length: 1024
Came here with length: 1024
Came here with length: 1024
Came here with length: 1024
Came here with length: 1024
Came here with length: 1024
Came here with length: 1024
Came here with length: 1024
Came here with length: 1024
Came here with length: 531

---

因此，对于小文件，它似乎在寻找download.gsp视图。但它不应该寻求这种观点，因为我在回复中加入了一些东西

我认为约书亚·摩尔的评论可能是正确的答案

但是，我建议稍微重构代码，使其更干净一点

1首先，尽量避免将与web相关的对象（如请求、响应）传递到服务层。您可以通过让服务层方法返回OutputStream来重构它。在服务方法中，您只需创建一个ByteArrayOutputStream，它将保存从S3下载的文件的内容。然后将该输出流的内容写入控制器内的response.outputStream。这将使测试和保持代码干燥变得更容易，也就是说，您现在可以使用此方法从S3下载文件，而无需将文件内容写入响应

2另外，您应该重命名控制器方法中声明的请求变量，因为请求是一个隐式变量

我的服务

霉菌控制者

---

它是如何失败的？提出了哪些例外情况？你犯了什么错误？您需要提供更多详细信息。@Joshuamore我已更新了更多详细信息。请在从该方法返回之前尝试outputStream.flush。

def amazonWebService
OutputStream downloadFileFromS3(String filename) {
    InputStream inputStream = amazonWebService.getS3("us-east-1").getObject(getBucketName(), filename).getObjectContent()
    OutputStream outputStream = new ByteArrayOutputStream()
    byte[] buffer = new byte[1024];
    int length
    log.info("This is the length: " + length)
    while ((length = inputStream.read(buffer)) != -1) {
        log.info("Came here with length: " + length)
        outputStream.write(buffer, 0, length)
    }
    if (inputStream) inputStream.close()
    return outputStream
}

    def myService
    def download() {

        // NOTE: I would recommend not using `request` as a variable 
        // name here since it's already an implicit variable 
        Request thatOtherRequest = Request.get(params.int("id")) 

        def filename = thatOtherRequest.origFileName
        def outputStream = myService.downloadFileFromS3UsingAwsSdk(filename)

        response.setContentType("application/octet-stream")
        response.setHeader("Content-disposition", "attachment;filename=${filename}")
        response.outputStream << outputStream
        response.outputStream.flush()
        return
    }