8. Topic Generation and Discovery

A DDS Topic represents the unit for information that can be produced or consumed by a DDS application. Topics are defined by a name, a type, and a set of QoS policies.

The Node.js DCPS API provides several ways of generating Node.js classes to represent DDS topics.

• over the wire discovery

• dynamic generation of Node.js classes using parameters IDL file and topic name

Note

• The Examples section provides the examples directory location, example descriptions and running instructions.

8.1. Over the Wire Discovery

Node.js topic classes can be generated for existing DDS topics in the DDS system. These topics are “discovered over the wire”.

The Node.js classes are generated when the topic is requested by name.

A code snippet is provided from findTopicExample.js. This example finds a topic registered by another process, and writes a sample to that topic.

Example: findTopic

const dds = require('vortexdds');
const dp = someAlreadyCreatedParticipant;
dp.findTopic('MyTopic')
  .then(topic => {
    // do something with the found topic
})
.catch(err => {
    // process any exception
})
.then(_ => {
    // cleanup: do any processing necessary after both
    // success and failure
});

8.2. Dynamic Generation of Node.js Topic Classes Using IDL and Name

The Node.js DCPS API supports generation of Node.js topic classes from IDL. This section describes the details of the IDL-Node.js binding.

8.2.1. Dynamic Generation

The Node.js DCPS API provides an asynchronous function that returns a Map of TypeSupport objects.

A TypeSupport object includes the topic typename, keys and descriptor.

The structure type representation of a topic is created by the TypeSupport object. However, the usage of the structure type is internal to the Node.js DCPS API.

In order to create a Topic, a topic name and a TypeSupport are passed into the Participant.createTopic function. (Qos and listener parameters are optional)

Example: Import a Topic from IDL

const dds = require('vortexdds');
const path = require('path');

//...

async function createTopic(participant) {

  const topicName = 'IoTData';
  const idlName = 'dds_IoTData.idl';
  const idlPath = path.resolve(idlName);

  //wait for dds.importIDL to return a map of typeSupports
  let typeSupports = await dds.importIDL(idlPath);
  let typeSupport = typeSupports.get('DDS::IoT::IoTData');

  return participant.createTopic(topicName, typeSupport);
};

8.2.2. Generated Artifacts

The following table defines the Node.js artifacts generated from IDL concepts:

IDL Concept	Node.js Concept
module	N/A
enum	enum from npm ‘enum’ package
enum value	enum value
struct	object
field	object property
union	object (IoTValue from dds_IoTData.idl is the only supported union)

Datatype mappings

The following table shows the Node.js equivalents to IDL primitive types:

IDL Type	Node.js Type
boolean	Boolean
char	Number
octet	Number
short	Number
ushort	Number
long	Number
ulong	Number
long long	Number
ulong long	Number
float	Number
double	Number
string	String
wchar	Unsupported
wstring	Unsupported
any	Unsupported
long double	Unsupported

Implementing Arrays and Sequences in Node.js

Both IDL arrays and IDL sequences are mapped to JavaScript arrays.

8.3. Limitations of Node.js Support

The Node.js binding has the following limitations:

• Listener is not supported for the Participant, Publisher and Subscriber entities.

• Only the IoTValue union from dds_IoTData.idl is supported.

• JavaScript does not currently include standard support for 64-bit integer values. 64-bit integers with more than 53 bits of data are represented by String values to avoid loss of precision. If the value will fit inside a JavaScript Number without losing precision, a Number can be used, otherwise use a String. (Refer to IoTData example which demonstrates the usage and ranges for the unsigned and signed 64 bit integers within nodejs.)