Update to LangChain4j 0.34

Closes: #873

core/runtime/src/main/java/io/quarkiverse/langchain4j/runtime/aiservice/AiServiceMethodImplementationSupport.java

@@ -134,6 +134,13 @@ private static Object doImplement(AiServiceMethodCreateInfo methodCreateInfo, Ob
         Object memoryId = memoryId(methodCreateInfo, methodArgs, context.chatMemoryProvider != null);
         boolean needsMemorySeed = needsMemorySeed(context, memoryId); // we need to know figure this out before we add the system and user message
 
+        boolean hasMethodSpecificTools = methodCreateInfo.getToolClassNames() != null
+                && !methodCreateInfo.getToolClassNames().isEmpty();
+        List<ToolSpecification> toolSpecifications = hasMethodSpecificTools ? methodCreateInfo.getToolSpecifications()
+                : context.toolSpecifications;
+        Map<String, ToolExecutor> toolExecutors = hasMethodSpecificTools ? methodCreateInfo.getToolExecutors()
+                : context.toolExecutors;
+
         Type returnType = methodCreateInfo.getReturnType();
         AugmentationResult augmentationResult = null;
         if (context.retrievalAugmentor != null) {
@@ -163,7 +170,12 @@ public Flow.Publisher<?> apply(AugmentationResult ar) {
                                 GuardrailsSupport.invokeInputGuardrails(methodCreateInfo, (UserMessage) augmentedUserMessage,
                                         context.chatMemory(memoryId), ar);
                                 List<ChatMessage> messagesToSend = messagesToSend(augmentedUserMessage, needsMemorySeed);
-                                return Multi.createFrom().emitter(new MultiEmitterConsumer(messagesToSend, context, memoryId));
+                                return Multi.createFrom()
+                                        .emitter(new MultiEmitterConsumer(messagesToSend, toolSpecifications,
+                                                toolExecutors,
+                                                ar.contents(),
+                                                context,
+                                                memoryId));
                             }
 
                             private List<ChatMessage> messagesToSend(ChatMessage augmentedUserMessage,
@@ -205,27 +217,21 @@ private List<ChatMessage> messagesToSend(ChatMessage augmentedUserMessage,
         if (isTokenStream(returnType)) {
             // TODO Indicate the output guardrails cannot be used when streaming
             chatMemory.commit(); // for streaming cases, we really have to commit because all alternatives are worse
-            return new AiServiceTokenStream(messagesToSend, context, memoryId);
+            return new AiServiceTokenStream(messagesToSend, toolSpecifications, toolExecutors,
+                    (augmentationResult != null ? augmentationResult.contents() : null), context, memoryId);
         }
 
         if (isMulti(returnType)) {
             // TODO Indicate the output guardrails cannot be used when streaming
             chatMemory.commit(); // for streaming cases, we really have to commit because all alternatives are worse
-            return Multi.createFrom().emitter(new MultiEmitterConsumer(messagesToSend, context, memoryId));
+            return Multi.createFrom().emitter(new MultiEmitterConsumer(messagesToSend, toolSpecifications,
+                    toolExecutors, augmentationResult != null ? augmentationResult.contents() : null, context, memoryId));
         }
 
         Future<Moderation> moderationFuture = triggerModerationIfNeeded(context, methodCreateInfo, messagesToSend);
 
         log.debug("Attempting to obtain AI response");
 
-        List<ToolSpecification> toolSpecifications = context.toolSpecifications;
-        Map<String, ToolExecutor> toolExecutors = context.toolExecutors;
-        // override with method specific info
-        if (methodCreateInfo.getToolClassNames() != null && !methodCreateInfo.getToolClassNames().isEmpty()) {
-            toolSpecifications = methodCreateInfo.getToolSpecifications();
-            toolExecutors = methodCreateInfo.getToolExecutors();
-        }
-
         Response<AiMessage> response = toolSpecifications == null
                 ? context.chatModel.generate(messagesToSend)
                 : context.chatModel.generate(messagesToSend, toolSpecifications);
@@ -572,19 +578,29 @@ public interface Wrapper {
 
     private static class MultiEmitterConsumer implements Consumer<MultiEmitter<? super String>> {
         private final List<ChatMessage> messagesToSend;
+        private final List<ToolSpecification> toolSpecifications;
+        private final Map<String, ToolExecutor> toolExecutors;
+        private final List<dev.langchain4j.rag.content.Content> contents;
         private final QuarkusAiServiceContext context;
         private final Object memoryId;
 
-        public MultiEmitterConsumer(List<ChatMessage> messagesToSend, QuarkusAiServiceContext context,
+        public MultiEmitterConsumer(List<ChatMessage> messagesToSend,
+                List<ToolSpecification> toolSpecifications,
+                Map<String, ToolExecutor> toolExecutors,
+                List<dev.langchain4j.rag.content.Content> contents,
+                QuarkusAiServiceContext context,
                 Object memoryId) {
             this.messagesToSend = messagesToSend;
+            this.toolSpecifications = toolSpecifications;
+            this.toolExecutors = toolExecutors;
+            this.contents = contents;
             this.context = context;
             this.memoryId = memoryId;
         }
 
         @Override
         public void accept(MultiEmitter<? super String> em) {
-            new AiServiceTokenStream(messagesToSend, context, memoryId)
+            new AiServiceTokenStream(messagesToSend, toolSpecifications, toolExecutors, contents, context, memoryId)
                     .onNext(em::emit)
                     .onComplete(new Consumer<>() {
                         @Override

docs/modules/ROOT/pages/includes/attributes.adoc

@@ -1,3 +1,3 @@
 :project-version: 0.17.2
-:langchain4j-version: 0.33.0
+:langchain4j-version: 0.34.0
 :examples-dir: ./../examples/

embedding-stores/pgvector/deployment/src/test/java/io/quarkiverse/langchain4j/pgvector/test/LangChain4jPgVectorBaseTest.java

@@ -65,7 +65,6 @@ void testSearch() {
             Embedding matchingEmbedding = this.embeddingModel().embed(textSegment).content();
             this.embeddingStore().add(matchingEmbedding, textSegment);
         });
-        this.awaitUntilPersisted();
         long totalIngestTime = System.currentTimeMillis() - startTime;
         Log.debugf("End Ingesting %s embeddings in %d ms.", nbRows, totalIngestTime);
         Embedding queryEmbedding = this.embeddingModel().embed("matching").content();

embedding-stores/pinecone/deployment/src/test/java/io/quarkiverse/langchain4j/pinecone/deployment/PineconeEmbeddingStoreTest.java

@@ -88,11 +88,6 @@ protected EmbeddingModel embeddingModel() {
         return embeddingModel;
     }
 
-    @Override
-    protected void awaitUntilPersisted() {
-        delay();
-    }
-
     @Override
     protected void clearStore() {
         Log.info("About to delete all embeddings");

model-providers/openai/azure-openai/runtime/src/main/java/io/quarkiverse/langchain4j/azure/openai/AzureOpenAiChatModel.java

@@ -15,6 +15,7 @@
 import java.time.Duration;
 import java.util.Collections;
 import java.util.List;
+import java.util.Locale;
 import java.util.Map;
 import java.util.concurrent.ConcurrentHashMap;
 
@@ -23,6 +24,8 @@
 import dev.ai4j.openai4j.OpenAiClient;
 import dev.ai4j.openai4j.chat.ChatCompletionRequest;
 import dev.ai4j.openai4j.chat.ChatCompletionResponse;
+import dev.ai4j.openai4j.chat.ResponseFormat;
+import dev.ai4j.openai4j.chat.ResponseFormatType;
 import dev.langchain4j.agent.tool.ToolSpecification;
 import dev.langchain4j.data.message.AiMessage;
 import dev.langchain4j.data.message.ChatMessage;
@@ -69,7 +72,7 @@ public class AzureOpenAiChatModel implements ChatLanguageModel, TokenCountEstima
     private final Double frequencyPenalty;
     private final Integer maxRetries;
     private final Tokenizer tokenizer;
-    private final String responseFormat;
+    private final ResponseFormat responseFormat;
     private final List<ChatModelListener> listeners;
 
     public AzureOpenAiChatModel(String endpoint,
@@ -119,7 +122,11 @@ public AzureOpenAiChatModel(String endpoint,
             throw new IllegalArgumentException("max-retries must be at least 1");
         }
         this.tokenizer = tokenizer;
-        this.responseFormat = responseFormat;
+        this.responseFormat = responseFormat == null ? null
+                : ResponseFormat.builder()
+                        .type(ResponseFormatType.valueOf(responseFormat.toUpperCase(Locale.ROOT)))
+                        .build();
+        ;
     }
 
     @Override

model-providers/openai/azure-openai/runtime/src/main/java/io/quarkiverse/langchain4j/azure/openai/AzureOpenAiImageModel.java

@@ -25,6 +25,7 @@
 import dev.ai4j.openai4j.SyncOrAsync;
 import dev.ai4j.openai4j.image.GenerateImagesRequest;
 import dev.ai4j.openai4j.image.GenerateImagesResponse;
+import dev.ai4j.openai4j.image.ImageData;
 import dev.langchain4j.data.image.Image;
 import dev.langchain4j.model.image.ImageModel;
 import dev.langchain4j.model.output.Response;
@@ -106,7 +107,7 @@ private void persistIfNecessary(GenerateImagesResponse response) {
             throw new UncheckedIOException(e);
         }
 
-        for (GenerateImagesResponse.ImageData data : response.data()) {
+        for (ImageData data : response.data()) {
             try {
                 data.url(
                         data.url() != null
@@ -118,7 +119,7 @@ private void persistIfNecessary(GenerateImagesResponse response) {
         }
     }
 
-    private static Image fromImageData(GenerateImagesResponse.ImageData data) {
+    private static Image fromImageData(ImageData data) {
         return Image.builder().url(data.url()).base64Data(data.b64Json()).revisedPrompt(data.revisedPrompt()).build();
     }
 

model-providers/openai/azure-openai/runtime/src/main/java/io/quarkiverse/langchain4j/azure/openai/AzureOpenAiStreamingChatModel.java

@@ -12,12 +12,15 @@
 import java.net.Proxy;
 import java.time.Duration;
 import java.util.List;
+import java.util.Locale;
 
 import dev.ai4j.openai4j.OpenAiClient;
 import dev.ai4j.openai4j.chat.ChatCompletionChoice;
 import dev.ai4j.openai4j.chat.ChatCompletionRequest;
 import dev.ai4j.openai4j.chat.ChatCompletionResponse;
 import dev.ai4j.openai4j.chat.Delta;
+import dev.ai4j.openai4j.chat.ResponseFormat;
+import dev.ai4j.openai4j.chat.ResponseFormatType;
 import dev.langchain4j.agent.tool.ToolSpecification;
 import dev.langchain4j.data.message.AiMessage;
 import dev.langchain4j.data.message.ChatMessage;
@@ -58,7 +61,7 @@ public class AzureOpenAiStreamingChatModel implements StreamingChatLanguageModel
     private final Double presencePenalty;
     private final Double frequencyPenalty;
     private final Tokenizer tokenizer;
-    private final String responseFormat;
+    private final ResponseFormat responseFormat;
 
     public AzureOpenAiStreamingChatModel(String endpoint,
             String apiVersion,
@@ -100,7 +103,11 @@ public AzureOpenAiStreamingChatModel(String endpoint,
         this.presencePenalty = presencePenalty;
         this.frequencyPenalty = frequencyPenalty;
         this.tokenizer = tokenizer;
-        this.responseFormat = responseFormat;
+        this.responseFormat = responseFormat == null ? null
+                : ResponseFormat.builder()
+                        .type(ResponseFormatType.valueOf(responseFormat.toUpperCase(Locale.ROOT)))
+                        .build();
+        ;
     }
 
     @Override