/* * Copyright 2004-2012 the original author or authors. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.springframework.webflow.execution; /** * A command that executes a behavior and returns a logical execution result a calling flow execution can respond to. * <p> * Actions typically delegate down to the application (or service) layer to perform business operations. They often * retrieve data to support response rendering. They act as a bridge between a SWF web-tier and your middle-tier * business logic layer. * <p> * When an action completes execution it signals a result event describing the outcome of that execution (for example, * "success", "error", "yes", "no", "tryAgain", etc). In addition to providing a logical outcome the flow can respond * to, a result event may have payload associated with it, for example a "success" return value or an "error" error * code. The result event is typically used as grounds for a state transition out of the current state of the calling * Flow. * <p> * Action implementations are often application-scoped singletons instantiated and managed by a web-tier Spring * application context to take advantage of Spring's externalized configuration and dependency injection capabilities * (which is a form of Inversion of Control [IoC]). Actions may also be stateful prototypes, storing conversational * state as instance variables. Action instance definitions may also be locally scoped to a specific flow definition * (see use of the "import" element of the root XML flow definition element.) * <p> * Note: Actions are directly instantiatable for use in a standalone test environment and can be parameterized with * mocks or stubs, as they are simple POJOs. Action proxies may also be generated at runtime for delegating to POJO * business operations that have no dependency on the Spring Web Flow API. * <p> * Note: if an Action is a singleton managed in application scope, take care not to store and/or modify caller-specific * state in a unsafe manner. The Action {@link #execute(RequestContext)} method runs in an independently executing * thread on each invocation so make sure you deal only with local data or internal, thread-safe services. * <p> * Note: an Action is not a controller like a Spring MVC controller or a Struts action is a controller. Flow actions are * <i>commands</i>. Such commands do not select views, they execute arbitrary behavioral logic and then return an * logical execution result. The flow that invokes an Action is responsible for responding to the execution result to * decide what to do next. In Spring Web Flow, the flow <i>is</i> the controller. * * @author Keith Donald * @author Erwin Vervaet */ public interface Action { /** * Execute this action. Action execution will occur in the context of a request associated with an active flow * execution. * <p> * Action invocation is typically triggered in a production environment by a state within a flow carrying out the * execution of a flow definition. The result of action execution, a logical outcome event, can be used as grounds * for a transition out of the calling state. * <p> * Note: The {@link RequestContext} argument to this method provides access to data about the active flow execution * in the context of the currently executing thread. Among other things, this allows this action to access * {@link RequestContext#getRequestScope() data} set by other actions, as well as set its own attributes it wishes * to expose in a given scope. * <p> * Some notes about actions and their usage of the attribute scope types: * <ul> * <li>Attributes set in {@link RequestContext#getRequestScope() request scope} exist for the life of the currently * executing request only. * <li>Attributes set in {@link RequestContext#getFlashScope() flash scope} exist until after view rendering is * completed. That time includes the current request plus any redirect required for the view render to complete. * <li>Attributes set in {@link RequestContext#getFlowScope() flow scope} exist for the life of the flow session and * will be cleaned up automatically when the flow session ends. * <li>Attributes set in {@link RequestContext#getConversationScope() conversation scope} exist for the life of the * entire flow execution representing a single logical "conversation" with a user. * </ul> * <p> * All attributes present in any scope are typically exposed in a model for access by a view when an "interactive" * state type such as a view state is entered. * <p> * Note: flow scope should generally not be used as a general purpose cache, but rather as a context for data needed * locally by other states of the flow this action participates in. For example, it would be inappropriate to stuff * large collections of objects (like those returned to support a search results view) into flow scope. Instead, put * such result collections in request scope, and ensure you execute this action again each time you wish to view * those results. 2nd level caches managed outside of SWF are more general cache solutions. * <p> * Note: as flow scoped attributes are eligible for serialization they should be <code>Serializable</code>. * * @param context the action execution context, for accessing and setting data in a {@link ScopeType scope type}, as * well as obtaining other flow contextual information (e.g. request context attributes and flow execution context * information) * @return a logical result outcome, used as grounds for a transition in the calling flow (e.g. "success", "error", * "yes", "no", * ...) * @throws Exception a exception occurred during action execution, either checked or unchecked; note, any * <i>recoverable</i> exceptions should be caught within this method and an appropriate result outcome returned * <i>or</i> be handled by the current state of the calling flow execution. */ public Event execute(RequestContext context) throws Exception; }