Skip to main content

Home/ Java Development/ Group items tagged rpc

Rss Feed Group items tagged

Filter: All | Bookmarks | Topics Simple Middle
Hendy Irawan

twitter/finagle @ GitHub - 0 views

twitter.github.com/finagle

twitter rpc scala java

shared by Hendy Irawan on 31 Oct 11 - No Cached
  • Hendy Irawan on 31 Oct 11
     
    "Finagle is a library for building asynchronous RPC servers and clients in Java, Scala, or any JVM language. Overview Built atop Netty, Finagle provides a rich set of tools that are protocol independent. Finagle is flexible enough to support a variety of RPC styles, including request-response, streaming, and pipelining (e.g., HTTP pipelining and Redis pipelining). It also makes it easy to work with stateful RPC styles (e.g., those requiring authentication and those that support transactions)."
anonymous

Getting Started with RequestFactory - Google Web Toolkit - Google Code - 0 views

code.google.com/...DevGuideRequestFactory.html

shared by anonymous on 10 Jan 11 - No Cached
  • Entity Proxies
    • anonymous
      anonymous on 10 Jan 11
       
      Proxy type (on the Client) vs Entity type (on the server)
  • proxy types
  • entity types
  • ...147 more annotations...
  • methods that return service stubs
  • one RequestFactory interface for your application
  • employeeRequest();
  • @Service(Employee.class)
  • extends RequestContext
  • extends RequestFactory
  • service stub
  • RequestFactory service stubs
  • must extend RequestContext
  • The methods in a service stub do not return entities directly
  • return subclasses of com.google.gwt.requestfactory.shared.Request.
  • This allows the methods on the interface to be invoked asynchronously with
  • Request.fire()
  • fire(    new Receiver()
  • onSuccess
  • callers pass an AsyncCallback that implements onSuccess()
  • takes a Receiver which must implement onSuccess()
  • Receiver is an abstract class having a default implementation of onFailure()
  • you can extend Receiver and override onFailure()
  • onViolation()
  • any constraint violations on the server
  • The Request type returned from each method
  • parameterized with the return type of the service method.
  • Methods that have no return value should return type Request<Void>
  • BigDecimal, BigInteger, Boolean, Byte, Enum, Character, Date, Double, Float, Integer, Long, Short, String, Void
  • subclass of EntityProxy
  • List<T> or Set<T>
  • primitive types are not supported
  • methods that operate on an entity itself
  • like persist() and remove()
  • return objects of type InstanceRequest rather than Reques
  • Server Implementations
  • methods defined in an
  • entity's service interface
  • implemented in the class named
  • @Service annotation
  • in these examples, is the entity class
  • service implementations do not directly implement the RequestContext interface
  • server-side implementations use the domain entity types
  • @Entity
  • EntityManager
  • createQuery
  • getResultList();
  • entityManager()
  • createEntityManager()
  • em.persist(this);
  • em.remove(attached
  • em.close();
  • defined in the service's
  • RequestContext interface
  • even though the implementation does not formally implement the interface in Java
  • name and argument list for each method
  • same on client and server
  • Client side methods
  • return Request<T>
  • only T on the server
  • EntityProxy types become the domain entity type on the server
  • Methods that return a Request object in the client interface are implemented as static methods on the entity
  • Methods that operate on a single instance of an entity, like persist() and remove(),
  • eturn an
  • InstanceRequest
  • in the client interface
  • Instance methods do not pass the instance directly, but rather via the
  • using()
  • instance methods must be implemented as non-static methods in the entity type
  • Four special methods are required on all entities
  • as they are used by the RequestFactory servlet:
  • constructor
  • findEntity
  • An entity's getId()
  • is typically auto-generated by the persistence engine (JDO, JPA, Objectify, etc.)
  • "find by ID" method has a special naming convention
  • find()
  • "find" plus the type's simple name
  • On the server
  • getVersion() method is used by RequestFactory to infer if an entity has changed
  • backing store (JDO, JPA, etc.) is responsible for updating the version each time the object is persisted,
  • RequestFactoryServlet sends an UPDATE
  • if an entity changes as
  • Second, the client maintains a version cache of recently seen entities
  • Whenever it sees an entity whose version has changed, it fires
  • UPDATE events on the event bus
  • so that listeners can update the view
  • GWT.create
  • and initialize it with your application's EventBus
  • GWT.create
  • requestFactory.initialize
  • create a new entity on the client
  • EmployeeRequest request
  • EmployeeProxy newEmployee
  • All client-side code should use the EmployeeProxy
  • not the Employee entity itself
  • unlike GWT-RPC, where the same concrete type is used on both client and server
  • RequestFactory
  • designed to be used with an ORM layer like JDO or JPA
  • on the server
  • to build data-oriented (CRUD) apps with an ORM-like interface
  • on the client
  • easy to implement a data access layer
  • structure your server-side code in a data-centric way
  • GWT-RPC, which is service-oriented
  • On the client side, RequestFactory keeps track of objects that have been modified and sends only changes
  • lightweight network payloads
  • solid foundation for automatic batching and caching of requests in the future
  • RequestFactoryServlet
  • RequestFactory uses its own servlet
  • own protocol
  • not designed for general purpose services like GWT-RPC
  • implements its
  • It is designed specifically for implementing a persistence layer on both client and server.
  • In persistence frameworks like JDO and JPA, entities are annotated with
  • client-side representation of an entity
  • known as a
  • DTO (Data Transfer Object)
  • hook used to indicate that an object can be managed by RequestFactory
  • RequestFactory
  • EntityProxy interface
  • automatically populates bean-style properties between entities on the server and the corresponding EntityProxy on the client,
  • send only changes ("deltas") to the server
  • extends EntityProxy
  • interface
  • @ProxyFor
  • reference the server-side entity being represented
  • It is not necessary to represent every property and method from the server-side entity in the EntityProxy
  • EntityProxyId returned by this method is used throughout RequestFactory-related classes
  • while getId() is shown in this example, most client code will want to refer to
  • EntityProxy.stableId() i
  • to represent any type
  • is not required to expose an ID and version
  • often used to represent embedded object types within entities
  • @Embedded
  • Address
  • Address type
  • POJO with no persistence annotations
  • Address is represented as a ValueProxy
  • extends ValueProxy
  • interface
  • extends EntityProxy
  • interface
  • AddressProxy
  • AddressProxy
  • ValueProxy can be used to pass any type to and from the server
  • RequestFactory
  • interface between your client and server code
  • RequestContext interface
  • The server-side service
  • must implement each method
anonymous

Organize Projects - Google Web Toolkit - Google Code - 0 views

code.google.com/...evGuideOrganizingProjects.html

shared by anonymous on 14 Jan 11 - Cached
  • com.google.gwt.gears.Gears
    • anonymous
      anonymous on 14 Jan 11
       
      Gears.gwt.xml does not define andy entry point. It can only be inherits
  • two ways to approach loading them
  • nclude each module with a separate <script> tag
  • ...118 more annotations...
  • Create a top level module XML definition
  • Compile the top level module
  • the second approach will lead to much better end-user performance
  • each module has to be downloaded separately by the end-user's browser
  • each module will contain redundant copies of GWT library
  • conflict with each other during event handling
  • Linkers are divided into three categories, PRE, POST, and PRIMARY
  • one primary linker is run for a compilation
  • everal linkers are provided by Core.gwt.xml, which is automatically inherited by User.gwt.xml.
  • monolithic JavaScript file.
  • cross-site deployment model.
  • standard iframe-based
  • <add-linker name="xs" />
  • The GWT compiler
  • packaging its output with the Linker subsystem
  • responsible for the final packaging of the JavaScript code
  • providing a pluggable bootstrap mechanism
  • re-use an existing Java API for a GWT project,
  • <super-source>
  • "re-root" a source path
  • to emulate part of the JRE not implemented by GWT
  • tells the compiler to add all subfolders of com/example/myproject/jre/
  • to the source path
  • com/google/myproject/gwt/jre/java/util/UUID.java
  • most commonly used elements in the module XML file.
  • <inherits name="
  • herits all the settings from the specified module
  • <entry-point class=
  • Entry points are all compiled into a single codebase
  • when the onModuleLoad() of your first entry point finishes, the next entry point is called immediately.
  • Any number of entry-point classes can be added
  • <source path="
  • resources get copied into the output directory during a GWT compile.
  • client subpackage is implicitly added to the source path
  • <public path="path" />
  • treated as a publicly-accessible resource.
  • resources get copied into the output directory
  • the public subpackage is implicitly added to the public
  • <servlet
  • For RPC, this element loads a servlet class
  • mounted at the specified URL path
  • path=
    • anonymous
      anonymous on 14 Jan 11
       
      monting location
  • class="
    • anonymous
      anonymous on 14 Jan 11
       
      Which servlet class
  • URL path should be absolute
  • @RemoteServiceRelativePath attribute
  • you must configure a WEB-INF/web.xml in your war directory to load any servlets needed.
  • n development mode,
  • <script src="
  • external JavaScrip
  • <stylesheet src="
  • Extends the set of values
  • for an existing client property
  • <replace-with-class>
  • <generate-with-class>
  • <when-property-is
  • three different types of predicates are
  • <when-type-assignable
  • <when-type-is
  • <all>
  • <any>
  • <none>
  • GWT libraries are organized into modules
  • you want to inherit at least the User module
  • contains all the core GWT functionality
  • including the EntryPoint class
  • widgets and panels
  • History feature
  • Internationalization
  • DOM programming, and more
  • Low-level HTTP
  • Use the following syntax to cause an external JavaScript file to be loaded
  • before your module entry point is called.
  • as if you had included it explicitly using the HTML <script
  • loaded before your onModuleLoad() is called.
  • all included scripts will be loaded when your application starts, in the order in which they are declared.
  • associate external CSS files with your module
  • GWT.getModuleBaseURL() + "foo.css" in client code
  • module's public path
  • useful when
  • inheritance makes resource inclusion particularly convenient.
  • If you wish to create a reusable library that relies upon particular stylesheets or JavaScript files, you can be sure that clients
  • see the documentation for FileSet for a general overview
  • <public>
  • <super-source>
  • <source>
  • includes
  • excludes
  • defaultexcludes
  • casesensitive
  • By default, the patterns listed here are excluded.
  • defaultexcludes is true
  • <script src='myApp/myApp.nocache.js'></script>
  • <script> tags always block evaluation of the page
  • <img> tags do not block page evaluation
  • two simultaneous connections
  • The body.onload() event will only fire once all external resources are fetched, including images and frames.
  • GWT selection script
  • like a normal script tag
  • but the compiled script will be fetched asynchronously.
  • Parsing is blocked until externalScriptZero.js is done fetching and evaluating.
  • myApp/myApp.nocache.js completes
  • the compiled scrip
  • (<hashname>.cache.html
  • begins fetching in a hidden IFRAME (this is non-blocking).
  • onModuleLoad() is not called yet, as we're still waiting on externalScriptOne.js
  • body.onload() fires
  • onload='alert("w00t!")
    • anonymous
      anonymous on 14 Jan 11
       
      is the last line executed
  • put the GWT selection script as early as possible
  • because it won't block any other script requests
  • <img> tags are not guaranteed to be done loading when onModuleLoad() is called
  • <script> tags are guaranteed to be done loading when onModuleLoad() is called
  • multiple EntryPoints
  • will all be called in sequence as soon as that module (and the outer document) is ready
  • multiple GWT modules within the same page
  • each module's EntryPoint will be called as soon as both that module and the outer document is ready
  • EntryPoints are not guaranteed to fire at the same time
  • or in the same order
  • in which their selection scripts were specified in the host page
1 - 3 of 3
Showing 20▼ items per page