@synefex/node-red-sila2

Node-RED nodes for talking to SiLA 2 lab-device servers from Node-RED. SiLA 2 is the open standard for lab automation, layering a Feature / Command / Property contract on top of gRPC over HTTP/2.

This package gives you a connection node and two action nodes built on @grpc/grpc-js and @grpc/proto-loader. Static proto loading (not reflection) so it works against every SiLA 2 server, regardless of implementation language.

Install

In your Node-RED userDir (typically ~/.node-red/):

npm install @synefex/node-red-sila2

Restart Node-RED. Three new nodes appear in the palette under the Synefex SiLA 2 section.

Nodes

| Node | Purpose | |---|---| | sila-connection | gRPC channel + bundled proto registry. Config node. | | sila-call-command | Invoke an unobservable SiLA 2 command on a feature. | | sila-get-property | Read an unobservable SiLA 2 property (Get_<Name>). |

Quick start

1. Drop a SiLA 2 connection config node and point it at your SiLA 2 server (host + port). Leave Insecure ticked for a dev server without TLS.
2. Wire an inject -> SiLA 2 get -> debug flow.
3. In the SiLA 2 get node:
   • Connection: the one you just made
   • Feature: SiLAService
   • Property: ServerName
4. Deploy and click inject. payload becomes the server's name string, e.g. "MettlerToledoAX205SiLAServer".

For commands, use SiLA 2 call instead. The bundled SiLAService feature exposes GetFeatureDefinition and SetServerName; for domain features (balances, liquid handlers, etc.) you bring your own .proto files (see "Adding more features" below).

Bundled features

The package always loads two .proto files from protos/:

• SiLAFramework.proto -- canonical wrapper types (String, Real, Boolean, ...) used by every SiLA 2 feature.
• SiLAService.proto -- the mandatory core feature every SiLA 2 server implements (ServerName, ServerType, ServerUUID, ServerDescription, ServerVersion, ServerVendorURL, ImplementedFeatures, plus GetFeatureDefinition and SetServerName). Hand-derived from the FDL in sila_base.

That's enough to talk to any SiLA 2 server's identity surface out of the box. For domain features (your actual lab devices) you provide their .proto files via the connection's Extra proto dirs / Extra proto files config.

Adding more features

SiLA 2 features are defined in FDL XML; the .proto is generated from the FDL by the SiLA 2 codegen. If your server is the sila2 Python reference implementation, the generated .proto files live inside the running container and can be copied out with one shell command (see the extract_protos.sh helper in this repo's sila-handoff/ notes).

Once you have the .proto files, drop them in a directory and point the connection node's Extra proto dirs at it (comma-separated list of dirs and/or individual files). The picker dropdowns automatically include any feature that's loaded.

Picker

Open a SiLA 2 call or SiLA 2 get node in the editor and you'll see two dropdowns above the manual fields:

• Feature: lists every loaded feature.
• Command / Property: lists the methods on the selected feature, classified by SiLA 2 naming convention (Get_<X> -> property <X>, everything else -> command).

The picker writes the chosen values into the manual text fields below; you can also type by hand.

Why static protos and not reflection

An earlier draft used gRPC reflection. The Python sila2 reference server enables reflection by default, but other SiLA 2 implementations (Java, C#) often do not, and even some sila2 deployments turn it off. Static proto loading is universal -- the package works against every SiLA 2 server we've tested.

Limitations / roadmap

This is the v1.0.0 release. Out of scope for now:

• Observable commands (the CommandExecutionUUID flow with progress).
• Observable properties (server-streaming subscriptions).
• Pinned-cert TLS (insecure and default-trust TLS are available).
• Decoded sila-error-bin trailer (SiLA 2 framework errors currently surface as generic gRPC errors).

If any of these is blocking for you, open an issue.

License

Apache License, Version 2.0. See LICENSE and NOTICE.