[Vertex AI] Make generateContentStream/sendMessageStream throws (#13573)

Author: andrewheard

FirebaseVertexAI/CHANGELOG.md

@@ -2,8 +2,10 @@
 - [fixed] Resolved a decoding error for citations without a `uri` and added
   support for decoding `title` fields, which were previously ignored. (#13518)
 - [changed] **Breaking Change**: The methods for starting streaming requests
-  (`generateContentStream` and `sendMessageStream`) and creating a chat instance
-  (`startChat`) are now asynchronous and must be called with `await`. (#13545)
+  (`generateContentStream` and `sendMessageStream`) are now throwing and
+  asynchronous and must be called with `try await`. (#13545, #13573)
+- [changed] **Breaking Change**: Creating a chat instance (`startChat`) is now
+  asynchronous and must be called with `await`. (#13545)
 
 # 10.29.0
 - [feature] Added community support for watchOS. (#13215)

FirebaseVertexAI/Sample/ChatSample/ViewModels/ConversationViewModel.swift

@@ -85,7 +85,7 @@ class ConversationViewModel: ObservableObject {
         guard let chat else {
           throw ChatError.notInitialized
         }
-        let responseStream = await chat.sendMessageStream(text)
+        let responseStream = try await chat.sendMessageStream(text)
         for try await chunk in responseStream {
           messages[messages.count - 1].pending = false
           if let text = chunk.text {

FirebaseVertexAI/Sample/FunctionCallingSample/ViewModels/FunctionCallingViewModel.swift

@@ -122,12 +122,12 @@ class FunctionCallingViewModel: ObservableObject {
     }
     let responseStream: AsyncThrowingStream<GenerateContentResponse, Error>
     if functionResponses.isEmpty {
-      responseStream = await chat.sendMessageStream(text)
+      responseStream = try await chat.sendMessageStream(text)
     } else {
       for functionResponse in functionResponses {
         messages.insert(functionResponse.chatMessage(), at: messages.count - 1)
       }
-      responseStream = await chat.sendMessageStream(functionResponses.modelContent())
+      responseStream = try await chat.sendMessageStream(functionResponses.modelContent())
     }
     for try await chunk in responseStream {
       processResponseContent(content: chunk)

FirebaseVertexAI/Sample/GenerativeAIMultimodalSample/ViewModels/PhotoReasoningViewModel.swift

@@ -84,7 +84,7 @@ class PhotoReasoningViewModel: ObservableObject {
         }
       }
 
-      let outputContentStream = await model.generateContentStream(prompt, images)
+      let outputContentStream = try await model.generateContentStream(prompt, images)
 
       // stream response
       for try await outputContent in outputContentStream {

FirebaseVertexAI/Sample/GenerativeAITextSample/ViewModels/SummarizeViewModel.swift

@@ -50,7 +50,7 @@ class SummarizeViewModel: ObservableObject {
 
       let prompt = "Summarize the following text for me: \(inputText)"
 
-      let outputContentStream = await model.generateContentStream(prompt)
+      let outputContentStream = try await model.generateContentStream(prompt)
 
       // stream response
       for try await outputContent in outputContentStream {

FirebaseVertexAI/Sources/Chat.swift

@@ -85,7 +85,7 @@ public actor Chat {
   /// - Parameter parts: The new content to send as a single chat message.
   /// - Returns: A stream containing the model's response or an error if an error occurred.
   @available(macOS 12.0, *)
-  public func sendMessageStream(_ parts: any ThrowingPartsRepresentable...)
+  public func sendMessageStream(_ parts: any ThrowingPartsRepresentable...) throws
     -> AsyncThrowingStream<GenerateContentResponse, Error> {
     return try sendMessageStream([ModelContent(parts: parts)])
   }
@@ -95,21 +95,16 @@ public actor Chat {
   /// - Parameter content: The new content to send as a single chat message.
   /// - Returns: A stream containing the model's response or an error if an error occurred.
   @available(macOS 12.0, *)
-  public func sendMessageStream(_ content: @autoclosure () throws -> [ModelContent])
+  public func sendMessageStream(_ content: @autoclosure () throws -> [ModelContent]) throws
     -> AsyncThrowingStream<GenerateContentResponse, Error> {
     let resolvedContent: [ModelContent]
     do {
       resolvedContent = try content()
     } catch let underlying {
-      return AsyncThrowingStream { continuation in
-        let error: Error
-        if let contentError = underlying as? ImageConversionError {
-          error = GenerateContentError.promptImageContentError(underlying: contentError)
-        } else {
-          error = GenerateContentError.internalError(underlying: underlying)
-        }
-        continuation.finish(throwing: error)
+      if let contentError = underlying as? ImageConversionError {
+        throw GenerateContentError.promptImageContentError(underlying: contentError)
       }
+      throw GenerateContentError.internalError(underlying: underlying)
     }
 
     return AsyncThrowingStream { continuation in
@@ -121,7 +116,7 @@ public actor Chat {
 
         // Send the history alongside the new message as context.
         let request = history + newContent
-        let stream = await model.generateContentStream(request)
+        let stream = try await model.generateContentStream(request)
         do {
           for try await chunk in stream {
             // Capture any content that's streaming. This should be populated if there's no error.

FirebaseVertexAI/Sources/GenerativeModel.swift

@@ -179,7 +179,7 @@ public final actor GenerativeModel {
   /// - Returns: A stream wrapping content generated by the model or a ``GenerateContentError``
   ///     error if an error occurred.
   @available(macOS 12.0, *)
-  public func generateContentStream(_ parts: any ThrowingPartsRepresentable...)
+  public func generateContentStream(_ parts: any ThrowingPartsRepresentable...) throws
     -> AsyncThrowingStream<GenerateContentResponse, Error> {
     return try generateContentStream([ModelContent(parts: parts)])
   }
@@ -190,21 +190,16 @@ public final actor GenerativeModel {
   /// - Returns: A stream wrapping content generated by the model or a ``GenerateContentError``
   ///     error if an error occurred.
   @available(macOS 12.0, *)
-  public func generateContentStream(_ content: @autoclosure () throws -> [ModelContent])
+  public func generateContentStream(_ content: @autoclosure () throws -> [ModelContent]) throws
     -> AsyncThrowingStream<GenerateContentResponse, Error> {
     let evaluatedContent: [ModelContent]
     do {
       evaluatedContent = try content()
     } catch let underlying {
-      return AsyncThrowingStream { continuation in
-        let error: Error
-        if let contentError = underlying as? ImageConversionError {
-          error = GenerateContentError.promptImageContentError(underlying: contentError)
-        } else {
-          error = GenerateContentError.internalError(underlying: underlying)
-        }
-        continuation.finish(throwing: error)
+      if let contentError = underlying as? ImageConversionError {
+        throw GenerateContentError.promptImageContentError(underlying: contentError)
       }
+      throw GenerateContentError.internalError(underlying: underlying)
     }
 
     let generateContentRequest = GenerateContentRequest(model: modelResourceName,

FirebaseVertexAI/Tests/Unit/ChatTests.swift

@@ -64,7 +64,7 @@ final class ChatTests: XCTestCase {
     )
     let chat = Chat(model: model, history: [])
     let input = "Test input"
-    let stream = await chat.sendMessageStream(input)
+    let stream = try await chat.sendMessageStream(input)
 
     // Ensure the values are parsed correctly
     for try await value in stream {

FirebaseVertexAI/Tests/Unit/GenerativeModelTests.swift

@@ -760,7 +760,7 @@ final class GenerativeModelTests: XCTestCase {
       )
 
     do {
-      let stream = await model.generateContentStream("Hi")
+      let stream = try await model.generateContentStream("Hi")
       for try await _ in stream {
         XCTFail("No content is there, this shouldn't happen.")
       }
@@ -784,7 +784,7 @@ final class GenerativeModelTests: XCTestCase {
       )
 
     do {
-      let stream = await model.generateContentStream(testPrompt)
+      let stream = try await model.generateContentStream(testPrompt)
       for try await _ in stream {
         XCTFail("No content is there, this shouldn't happen.")
       }
@@ -807,7 +807,7 @@ final class GenerativeModelTests: XCTestCase {
       )
 
     do {
-      let stream = await model.generateContentStream("Hi")
+      let stream = try await model.generateContentStream("Hi")
       for try await _ in stream {
         XCTFail("No content is there, this shouldn't happen.")
       }
@@ -827,7 +827,7 @@ final class GenerativeModelTests: XCTestCase {
       )
 
     do {
-      let stream = await model.generateContentStream("Hi")
+      let stream = try await model.generateContentStream("Hi")
       for try await _ in stream {
         XCTFail("Content shouldn't be shown, this shouldn't happen.")
       }
@@ -847,7 +847,7 @@ final class GenerativeModelTests: XCTestCase {
       )
 
     do {
-      let stream = await model.generateContentStream("Hi")
+      let stream = try await model.generateContentStream("Hi")
       for try await _ in stream {
         XCTFail("Content shouldn't be shown, this shouldn't happen.")
       }
@@ -866,7 +866,7 @@ final class GenerativeModelTests: XCTestCase {
         withExtension: "txt"
       )
 
-    let stream = await model.generateContentStream("Hi")
+    let stream = try await model.generateContentStream("Hi")
     do {
       for try await content in stream {
         XCTAssertNotNil(content.text)
@@ -887,7 +887,7 @@ final class GenerativeModelTests: XCTestCase {
       )
 
     var responses = 0
-    let stream = await model.generateContentStream("Hi")
+    let stream = try await model.generateContentStream("Hi")
     for try await content in stream {
       XCTAssertNotNil(content.text)
       responses += 1
@@ -904,7 +904,7 @@ final class GenerativeModelTests: XCTestCase {
       )
 
     var responses = 0
-    let stream = await model.generateContentStream("Hi")
+    let stream = try await model.generateContentStream("Hi")
     for try await content in stream {
       XCTAssertNotNil(content.text)
       responses += 1
@@ -921,7 +921,7 @@ final class GenerativeModelTests: XCTestCase {
       )
 
     var hadUnknown = false
-    let stream = await model.generateContentStream("Hi")
+    let stream = try await model.generateContentStream("Hi")
     for try await content in stream {
       XCTAssertNotNil(content.text)
       if let ratings = content.candidates.first?.safetyRatings,
@@ -940,7 +940,7 @@ final class GenerativeModelTests: XCTestCase {
         withExtension: "txt"
       )
 
-    let stream = await model.generateContentStream("Hi")
+    let stream = try await model.generateContentStream("Hi")
     var citations = [Citation]()
     var responses = [GenerateContentResponse]()
     for try await content in stream {
@@ -996,7 +996,7 @@ final class GenerativeModelTests: XCTestCase {
         appCheckToken: appCheckToken
       )
 
-    let stream = await model.generateContentStream(testPrompt)
+    let stream = try await model.generateContentStream(testPrompt)
     for try await _ in stream {}
   }
 
@@ -1018,7 +1018,7 @@ final class GenerativeModelTests: XCTestCase {
         appCheckToken: AppCheckInteropFake.placeholderTokenValue
       )
 
-    let stream = await model.generateContentStream(testPrompt)
+    let stream = try await model.generateContentStream(testPrompt)
     for try await _ in stream {}
   }
 
@@ -1030,7 +1030,7 @@ final class GenerativeModelTests: XCTestCase {
       )
     var responses = [GenerateContentResponse]()
 
-    let stream = await model.generateContentStream(testPrompt)
+    let stream = try await model.generateContentStream(testPrompt)
     for try await response in stream {
       responses.append(response)
     }
@@ -1056,7 +1056,7 @@ final class GenerativeModelTests: XCTestCase {
 
     var responseCount = 0
     do {
-      let stream = await model.generateContentStream("Hi")
+      let stream = try await model.generateContentStream("Hi")
       for try await content in stream {
         XCTAssertNotNil(content.text)
         responseCount += 1
@@ -1076,7 +1076,7 @@ final class GenerativeModelTests: XCTestCase {
   func testGenerateContentStream_nonHTTPResponse() async throws {
     MockURLProtocol.requestHandler = try nonHTTPRequestHandler()
 
-    let stream = await model.generateContentStream("Hi")
+    let stream = try await model.generateContentStream("Hi")
     do {
       for try await content in stream {
         XCTFail("Unexpected content in stream: \(content)")
@@ -1096,7 +1096,7 @@ final class GenerativeModelTests: XCTestCase {
         withExtension: "txt"
       )
 
-    let stream = await model.generateContentStream(testPrompt)
+    let stream = try await model.generateContentStream(testPrompt)
     do {
       for try await content in stream {
         XCTFail("Unexpected content in stream: \(content)")
@@ -1120,7 +1120,7 @@ final class GenerativeModelTests: XCTestCase {
         withExtension: "txt"
       )
 
-    let stream = await model.generateContentStream(testPrompt)
+    let stream = try await model.generateContentStream(testPrompt)
     do {
       for try await content in stream {
         XCTFail("Unexpected content in stream: \(content)")
@@ -1159,7 +1159,7 @@ final class GenerativeModelTests: XCTestCase {
     )
 
     var responses = 0
-    let stream = await model.generateContentStream(testPrompt)
+    let stream = try await model.generateContentStream(testPrompt)
     for try await content in stream {
       XCTAssertNotNil(content.text)
       responses += 1