超时时间超过了我在“java.net.URLConnection”中设置的时间
我正在尝试为 java.net.URLConnection 设置一个超时，如下所示：
URL url = new URL("http://google.com");
URLConnection con = url.openConnection();
con.setConnectTimeout(2000);
con.setReadTimeout(2000);
InputStream in = con.getInputStream();
doc = new Tidy().parseDOM(in, null);
XPath xpath = XPathFactory.newInstance().newXPath();
XPathExpression expr = xpath.compile("//td//span");
Node form = (Node) expr.evaluate(doc, XPathConstants.NODE);
问题是，当我将超时设置为 2000（2 秒）时，它等待的时间会超过它（例如 20 秒，因为我我自己统计的）直到抛出Timeout异常。
有什么问题吗？
如何将超时设置为完全按照我设置的值工作？
I'm trying to set a timeout for java.net.URLConnection as this:
URL url = new URL("http://google.com");
URLConnection con = url.openConnection();
con.setConnectTimeout(2000);
con.setReadTimeout(2000);
InputStream in = con.getInputStream();
doc = new Tidy().parseDOM(in, null);
XPath xpath = XPathFactory.newInstance().newXPath();
XPathExpression expr = xpath.compile("//td//span");
Node form = (Node) expr.evaluate(doc, XPathConstants.NODE);
The problem is that while I set timeout to 2000 (2 seconds) it waits more than it (for example 20 seconds as I counted by myself) until throwing Timeout exception.
What is the problem?
How Can I set the timeout to work exactly at the value I set?
如果你对这篇内容有疑问，欢迎到本站社区发帖提问 参与讨论，获取更多帮助，或者扫码二维码加入 Web 技术交流群。
绑定邮箱获取回复消息
由于您还没有绑定你的真实邮箱，如果其他用户或者作者回复了您的评论，将不能在第一时间通知您！
发布评论
评论（1）
readTimeOut在 Sun (Oracle) 的实现中无法可靠地工作。这甚至在 JavaDocs：不过，您可以使用线程或计时器来模拟超时。将连接传递给子线程，并要求该线程在超时到期后调用
disconnect或类似的操作。The
readTimeOutis not working in Sun's (Oracle's) implementation reliably. This is even mentioned in the JavaDocs:You can simulate timeouts using threads or timers though. Pass the connection to a child thread, and ask that thread to call
disconnector something similar after the timeout expires.