|Apache Struts 2 Documentation > Home > Guides > Core Developers Guide > Interceptors > Execute and Wait Interceptor|
The ExecuteAndWaitInterceptor is great for running long-lived actions in the background while showing the user a nice progress meter. This also prevents the HTTP request from timing out when the action takes more than 5 or 10 minutes.
Using this interceptor is pretty straight forward. Assuming that you are including struts-default.xml, this interceptor is already configured but is not part of any of the default stacks. Because of the nature of this interceptor, it must be the last interceptor in the stack.
This interceptor works on a per-session basis. That means that the same action name (myLongRunningAction, in the above example) cannot be run more than once at a time in a given session. On the initial request or any subsequent requests (before the action has completed), the wait result will be returned. The wait result is responsible for issuing a subsequent request back to the action, giving the effect of a self-updating progress meter.
If no "wait" result is found, Struts will automatically generate a wait result on the fly. This result is written in FreeMarker and cannot run unless FreeMarker is installed. If you don't wish to deploy with FreeMarker, you must provide your own wait result. This is generally a good thing to do anyway, as the default wait page is very plain.
Whenever the wait result is returned, the action that is currently running in the background will be placed on top of the stack. This allows you to display progress data, such as a count, in the wait page. By making the wait page automatically reload the request to the action (which will be short-circuited by the interceptor), you can give the appearance of an automatic progress meter.
This interceptor also supports using an initial wait delay. An initial delay is a time in milliseconds we let the
server wait before the wait page is shown to the user. During the wait this interceptor will wake every 100 millis
to check if the background process is done premature, thus if the job for some reason doesn't take to long the wait
page is not shown to the user.
This is useful for e.g. search actions that have a wide span of execution time. Using a delay time of 2000 millis we ensure the user is presented fast search results immediately and for the slow results a wait page is used.
Important: Because the action will be running in a seperate thread, you can't use ActionContext because it is a ThreadLocal. This means if you need to access, for example, session data, you need to implement SessionAware rather than calling ActionContext.getSession().
The thread kicked off by this interceptor will be named in the form actionNameBackgroundProcess. For example, the search action would run as a thread named searchBackgroundProcess.
waitas result code). Default is no initial delay.
If you wish to make special preparations before and/or after the invocation of the background thread, you can extend the BackgroundProcess class and implement the beforeInvocation() and afterInvocation() methods. This may be useful for obtaining and releasing resources that the background process will need to execute successfully. To use your background process extension, extend ExecuteAndWaitInterceptor and implement the getNewBackgroundProcess() method.