Hello,

You can use the browse methods of the AddressSpace instance accessible via the getAddressSpace method of UaClient instance. You can start browsing from the Root (i=84) or a specific Node by using the NodeId of that Node as a parameter of the browse method. You can then repeat this process for all the results of the browse to recursively traverse the entire AddressSpace.

The ReferenceDescriptions returned by the browse methods include the NodeClasses of the Nodes, so you can use that information to read Values of Variable Nodes you encounter while browsing. You use the com.prosysopc.ua.client.UaClient.readValue(NodeId) method to do this.

There are three things to consider when browsing the AddressSpace recursively:

First, the number of Nodes in the AddressSpace can be very large so traversing through all of them can take a long time. In general, I wouldn’t recommend starting from the Root as that would mean you’d also browse all the TypeDefinition Nodes. If you don’t need to browse the TypeDefinition Nodes, I would recommend starting from the Objects (i=85) instead. It is possible for the Objects to also contain hundreds or thousands of Nodes so it would be better to start from a specific Node that has a reasonable amount of “child” Nodes.

Second, you should limit the types of References returned by browse by using the overloads of the method that allow specifying a ReferenceType such as com.prosysopc.ua.client.AddressSpace.browse(NodeId, BrowseDirection, NodeId). Generally a good idea would be to limit the browse to HierarchicalReferences (i=33) unless you would need metadata such as TypeDefinitions of Nodes. Alternatively, you can get all References and choose to follow only some of them in the recursive browses.

Third, it is possible for the AddressSpace to contain loops where browsing from a Node would eventually lead to the same Node. Thus, you must keep track of the NodeIds of the Nodes you’ve already browsed to avoid getting stuck in an infinite loop.