* One or more {@link HttpHandler} objects must be associated with a server * in order to process requests. Each such HttpHandler is registered * with a root URI path which represents the * location of the application or service on this server. The mapping of a handler * to a HttpServer is encapsulated by a {@link HttpContext} object. HttpContexts * are created by calling {@link #createContext(String,HttpHandler)}. * Any request for which no handler can be found is rejected with a 404 response. * Management of threads can be done external to this object by providing a * {@link java.util.concurrent.Executor} object. If none is provided a default * implementation is used. *
* * Mapping request URIs to HttpContext paths
* When a HTTP request is received, * the appropriate HttpContext (and handler) is located by finding the context * whose path is the longest matching prefix of the request URI's path. * Paths are matched literally, which means that the strings are compared * case sensitively, and with no conversion to or from any encoded forms. * For example. Given a HttpServer with the following HttpContexts configured.
*
| Context | Context path |
| ctx1 | "/" |
| ctx2 | "/apps/" |
| ctx3 | "/apps/foo/" |
* the following table shows some request URIs and which, if any context they would * match with.
*
| Request URI | Matches context |
| "http://foo.com/apps/foo/bar" | ctx3 |
| "http://foo.com/apps/Foo/bar" | no match, wrong case |
| "http://foo.com/apps/app1" | ctx2 |
| "http://foo.com/foo" | ctx1 |
* Note about socket backlogs
* When binding to an address and port number, the application can also specify an integer
* backlog parameter. This represents the maximum number of incoming TCP connections
* which the system will queue internally. Connections are queued while they are waiting to
* be accepted by the HttpServer. When the limit is reached, further connections may be
* rejected (or possibly ignored) by the underlying TCP implementation. Setting the right
* backlog value is a compromise between efficient resource usage in the TCP layer (not setting
* it too high) and allowing adequate throughput of incoming requests (not setting it too low).
* @since 1.6
*/
@jdk.Exported
public abstract class HttpServer {
/**
*/
protected HttpServer () {
}
/**
* creates a HttpServer instance which is initially not bound to any local address/port.
* The HttpServer is acquired from the currently installed {@link HttpServerProvider}
* The server must be bound using {@link #bind(InetSocketAddress,int)} before it can be used.
* @throws IOException
*/
public static HttpServer create () throws IOException {
return create (null, 0);
}
/**
* Create a HttpServer instance which will bind to the
* specified {@link java.net.InetSocketAddress} (IP address and port number)
*
* A maximum backlog can also be specified. This is the maximum number of
* queued incoming connections to allow on the listening socket.
* Queued TCP connections exceeding this limit may be rejected by the TCP implementation.
* The HttpServer is acquired from the currently installed {@link HttpServerProvider}
*
* @param addr the address to listen on, if null then bind() must be called
* to set the address
* @param backlog the socket backlog. If this value is less than or equal to zero,
* then a system default value is used.
* @throws BindException if the server cannot bind to the requested address,
* or if the server is already bound.
* @throws IOException
*/
public static HttpServer create (
InetSocketAddress addr, int backlog
) throws IOException {
HttpServerProvider provider = HttpServerProvider.provider();
return provider.createHttpServer (addr, backlog);
}
/**
* Binds a currently unbound HttpServer to the given address and port number.
* A maximum backlog can also be specified. This is the maximum number of
* queued incoming connections to allow on the listening socket.
* Queued TCP connections exceeding this limit may be rejected by the TCP implementation.
* @param addr the address to listen on
* @param backlog the socket backlog. If this value is less than or equal to zero,
* then a system default value is used.
* @throws BindException if the server cannot bind to the requested address or if the server
* is already bound.
* @throws NullPointerException if addr is null
*/
public abstract void bind (InetSocketAddress addr, int backlog) throws IOException;
/**
* Starts this server in a new background thread. The background thread
* inherits the priority, thread group and context class loader
* of the caller.
*/
public abstract void start () ;
/**
* sets this server's {@link java.util.concurrent.Executor} object. An
* Executor must be established before {@link #start()} is called.
* All HTTP requests are handled in tasks given to the executor.
* If this method is not called (before start()) or if it is
* called with a null Executor, then
* a default implementation is used, which uses the thread
* which was created by the {@link #start()} method.
* @param executor the Executor to set, or null for default
* implementation
* @throws IllegalStateException if the server is already started
*/
public abstract void setExecutor (Executor executor);
/**
* returns this server's Executor object if one was specified with
* {@link #setExecutor(Executor)}, or null if none was
* specified.
* @return the Executor established for this server or null if not set.
*/
public abstract Executor getExecutor () ;
/**
* stops this server by closing the listening socket and disallowing
* any new exchanges from being processed. The method will then block
* until all current exchange handlers have completed or else when
* approximately delay seconds have elapsed (whichever happens
* sooner). Then, all open TCP connections are closed, the background
* thread created by start() exits, and the method returns.
* Once stopped, a HttpServer cannot be re-used.
* * @param delay the maximum time in seconds to wait until exchanges have finished. * @throws IllegalArgumentException if delay is less than zero. */ public abstract void stop (int delay); /** * Creates a HttpContext. A HttpContext represents a mapping from a * URI path to a exchange handler on this HttpServer. Once created, all requests * received by the server for the path will be handled by calling * the given handler object. The context is identified by the path, and * can later be removed from the server using this with the {@link #removeContext(String)} method. *
* The path specifies the root URI path for this context. The first character of path must be * '/'.
* The class overview describes how incoming request URIs are mapped
* to HttpContext instances.
* @param path the root URI path to associate the context with
* @param handler the handler to invoke for incoming requests.
* @throws IllegalArgumentException if path is invalid, or if a context
* already exists for this path
* @throws NullPointerException if either path, or handler are null
*/
public abstract HttpContext createContext (String path, HttpHandler handler) ;
/**
* Creates a HttpContext without initially specifying a handler. The handler must later be specified using
* {@link HttpContext#setHandler(HttpHandler)}. A HttpContext represents a mapping from a
* URI path to an exchange handler on this HttpServer. Once created, and when
* the handler has been set, all requests
* received by the server for the path will be handled by calling
* the handler object. The context is identified by the path, and
* can later be removed from the server using this with the {@link #removeContext(String)} method.
*
* The path specifies the root URI path for this context. The first character of path must be * '/'.
* The class overview describes how incoming request URIs are mapped
* to HttpContext instances.
* @param path the root URI path to associate the context with
* @throws IllegalArgumentException if path is invalid, or if a context
* already exists for this path
* @throws NullPointerException if path is null
*/
public abstract HttpContext createContext (String path) ;
/**
* Removes the context identified by the given path from the server.
* Removing a context does not affect exchanges currently being processed
* but prevents new ones from being accepted.
* @param path the path of the handler to remove
* @throws IllegalArgumentException if no handler corresponding to this
* path exists.
* @throws NullPointerException if path is null
*/
public abstract void removeContext (String path) throws IllegalArgumentException ;
/**
* Removes the given context from the server.
* Removing a context does not affect exchanges currently being processed
* but prevents new ones from being accepted.
* @param context the context to remove
* @throws NullPointerException if context is null
*/
public abstract void removeContext (HttpContext context) ;
/**
* returns the address this server is listening on
* @return the address/port number the server is listening on
*/
public abstract InetSocketAddress getAddress() ;
}