...
Applications work with messages as normal java objects. These views must be serialized into a provider specific message when sent, and when received they must be deserialized (or wrapped). The Talon Application Data Modeler generates Factory objects that facilitate The application registers the generated factories that it uses with the Talon runtime to allow deserialization of messages as they are received from the messaging provider.
Message Engine Independent Serializers
...
Code Block | ||||
---|---|---|---|---|
| ||||
@Generated(value="com.neeve.adm.AdmGenerator", date="Fri Jan 23 02:03:22 PST 2015") public interface IMyMessage extends extends IRogNode, IRogMessage { public final byte [] serializeToByteArray(); public final ByteBuffer serializeToByteBuffer(); public final String serializeToJson(); public final void deserializeFromByteArray(byte[] bytes); public final void deserializeFromByteBuffer(ByteBuffer bytes); public final void deserializeFromJson(String json) } |
Message Metatadata
Anchor | ||||
---|---|---|---|---|
|
When a message is transported over a bus binding, metadata is attached to the message to assist in routing, encoding/decoding, and de-duplication. The mechanism of how this metadata is attached to a message is specific to each bus binding implementation. For cases where you are working with the messaging provider directly (not through a bus binding implementation it useful to understand the structure of message metadata and how it is serialized.
Field | Type | Description |
---|---|---|
Channel Identification | ||
MessageChannelName | String | The name of the channel. This is used by the receiver to lookup the channel on which the message should be dispatched to the application. |
MessageChannelId | short | The id of the channel. This is used by the receiver to lookup the channel on which the message should be dispatched to the application. Using a channel id allows for more efficient lookup of the channel, but care must be taken that the specified channel id matches that configured for receivers of the message will be flagged as an unhandled message. |
Message Encoding Message encoding fields allows a receiving application to lookup a message factory and use it to deserialize or wrap a message. | ||
MessageViewFactory | short | The id of the factory that is used to deserialize the message. This is used by the receiver to lookup the factory that is used to deserialize (or wrap) the received message payload. The message view factory id can be determined by calling MessageView.getVfid()on a message. For ADM modeled messages, this will be the negative of the factory id defined in the model (user factory types are negative at runtime to distinguish between user and platform factory types). |
MessageViewType | short | The id of the type (unique to its factory) used to deserialize the message. The message view type can be determined by calling MessageView.getType() on a message. For ADM modeled messages, this will be the id of the message defined in the ADM model. |
MessageEncodingType | byte | The encoding type of the message. The encoding type is used by the factory to differentiate between possible encodings supported by the message, but may also be used by wire sniffers as a hint to decoding the message. Valid values are:
|
Message Sequencing Message sequencing fields allow for the receiving application to filter duplicate and out of order messages. For a given sender id and flow, an AepEngine will filter out received messages that don't have a monotonically increasing sequence number | ||
MessageSender | int | A globally unique id that identifies the sender of a message. By default an AepEngine uses the hashcode of the engine name as the sender id. |
MessageSno | long | A monotonically increasing sequence number.
|
MessageFlow | int | Indicates the flow to which the message belongs. Flows allow for partitioning of message traffic in conjunction with application state and allow intra application state partitioning. ![]() |
...