Post

Replies

Boosts

Views

Activity

LanguageModelSession with multiple tools and structured outpout
Hi, I'm using LanguageModelSession and giving it two different tools to query data from a local database. I'm wondering how I can have the session generate structured content as the response that includes data one or both tools (or no tool at all). Here is an example of what I'm trying to do: Let's say the app has access to a database that contains information about exercise and sleep data (this is just an analogy). There are two tools, GetExerciseData() and GetSleepData(). The user may then prompt something like, "how well did I sleep in November". I have this working so that it calls through to the right tool, which would return a SleepSummary. However, I can't figure out how to have the session return the right structured data. I can do this and get back good text data: let response = session.respond(to: userInput), but I believe I want to do something like: let response = session.respond(to: trimmed, generating: <SomeStructure?>) Sometimes the model I run one tool or the other, or both tools, or no tool at all. Any help of what the right way to go about this would be much appreciated. Most of the example I found have to do with 1 tool.
1
0
479
3w
Help with dates in Foundation Model custom Tool
I have an app that stores lots of data that is of interest to the user. Analogies would be the Photos apps or the Health app. I'm trying to use the Foundation Models framework to allow users to surface information they find interesting using natural language, for example, "Tell me about the widgets from yesterday" or "Tell me about the widgets for the last 3 days". Specifically, I'm trying to get a date range passed down to the Tool so that I can pull the relevant widgets from the database in the call function. What is the right way to set up the Arguments to get at a date range?
3
0
718
Dec ’25
Can't get a scoped resource URL from drag and drop
Hi, My Mac app allows a customer to drag and drop a file package onto a SwiftUI view. I can't seem to find a way to successfully call .startAccessingSecurityScopedResource() with the file/dir that was dropped into the view. I put together a simple test app. Here is the code: struct ContentView: View { @State var isTargetedForDrop: Bool = false var body: some View { VStack { Image(systemName: "globe") .imageScale(.large) .foregroundStyle(.tint) Text("Hello, world!") Rectangle() .stroke(Color.gray) .onDrop(of: [UTType.fileURL], isTargeted: $isTargetedForDrop) { providers in guard let provider = providers.first(where: { $0.hasItemConformingToTypeIdentifier(UTType.fileURL.identifier) }) else { return false } provider.loadItem(forTypeIdentifier: UTType.fileURL.identifier, options: nil) { item, error in if let error = error { print("Drop load error: \(error)") return } if let url = item as? URL { print("Dropped file URL: \(url)") } else if let data = item as? Data, let url = URL(dataRepresentation: data, relativeTo: nil) { print("Dropped file URL (from data): \(url)") let access = url.startAccessingSecurityScopedResource() if access { print("Successfully accessed file at URL: \(url)") } else { print("Failed to access file at URL: \(url)") } url.stopAccessingSecurityScopedResource() } else { print("Unsupported dropped item: \(String(describing: item))") } } return true } } .padding() } } When I drop a file package into this view I see, "Failed to access file at URL: <the_full_file_path>" I'm running Xcode 26 on macOS 26.
1
0
102
Oct ’25
Trouble loading ModelEntity from RealityKitContent package
Hi, I'm working on a simple visionOS app and I'm testing on device. For one part of the app, I load an object in and place it on the user's hand. If I use a primitive shape, like a sphere or cylinder, this works fine. However, now I'm trying to load a an object from my RealityKitContent package. But everytime I try this, I get a an error message, resourceNotFound("Stone"), where "Stone" is one of my usda scenes. This is what the guts of my function looks like that should return a ModelEntity: do { let entity = try await ModelEntity(named: "Stone", in: realityKitContentBundle) entity.generateCollisionShapes(recursive: true) return entity } catch { print("Error \(error)") } I can see the "Stone" in my Xcode sidebar as part of the RealityKitContent package and inside that scene, there is a simple sphere, but alas I always get this in the Xcode console, "Error resourceNotFound("Stone")" I'm probably doing something pretty silly, hopefully it's obvious to someone else. Thanks for the help. Ian
3
1
1.6k
Feb ’24