如何关闭AWS S3读取流(AWSJavaScriptSDK)

问题描述

我有一个AWS S3对象和在其上创建的读取流，如下所示:

I have an AWS S3 object and a read stream created on it like this:

const s3 = new AWS.S3();

const readStream = s3
  .getObject(params)
  .createReadStream()
  .on('error', err => {
    // do something
  });

现在，当流在120秒后未读到末尾(例如，流媒体被客户端中止)时，将触发 error 事件，发生以下情况: TimeoutError:120000ms后连接超时

Now when the stream is not read to the end (e.g. the streaming is aborted by client) after 120 sec the error event is triggered with: TimeoutError: Connection timed out after 120000ms

如何关闭流(或整个S3对象)?

How can I close the stream (or the entire S3 object)?

我尝试在此处中记录的 readStream.destroy()a>，但是不起作用.

I tried readStream.destroy() that is documented here, but it does not work.

推荐答案

我一直在寻找类似情况的解决方案，并碰到了这个线程.

I was looking for a solution to a similar case and bumped into this thread.

有一个 AWS请求中止此处记录的方法，该方法可让您在不接收所有数据的情况下取消请求(这与节点的http请求类似).

There is an AWS Request abort method documented here which allows you to cancel the request without receiving all the data (it's a similar concept to node's http request).

您的代码应如下所示:

const s3 = new AWS.S3();

const request = s3.getObject(params);
const readStream = request.createReadStream()
  .on('error', err => {
    request.abort(); // and do something else also...
  });

可能是错误的，但就我而言-我正在获取数据，并且要在达到特定点时停止流传输(即，在文件中找到特定数据，这只是检查它是否是唯一的问题)存在-我什么都不需要了.

It may be on error, but in my case - I'm fetching data and I want to stop streaming when I've reached a certain point (i.e. found specific data in the file and it's only a matter of checking if it exists - I don't need anything else).

以上内容也可以与 request 和 node-fetch 模块一起很好地工作.

The above will work well with both request and node-fetch modules as well.

这篇关于如何关闭AWS S3读取流(AWSJavaScriptSDK)的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持！