wip - working on timeout issue

2025-12-18 20:04:17 +01:00 · 2025-11-06 18:17:06 +00:00 · 2025-11-06 18:17:06 +00:00 · 3695a9a70a
commit 3695a9a70a
parent 98d4949930
13 changed files with 141 additions and 51 deletions
--- a/app/core/src/main/resources/application.properties
+++ b/app/core/src/main/resources/application.properties
@ -4,9 +4,8 @@ logging.level.org.springframework.security=WARN
 logging.level.org.hibernate=WARN
 logging.level.org.eclipse.jetty=WARN
 #logging.level.org.springframework.security.oauth2=DEBUG
-#logging.level.org.springframework.security=DEBUG
 #logging.level.org.opensaml=DEBUG
-#logging.level.stirling.software.proprietary.security=DEBUG
+logging.level.stirling.software.proprietary.security=DEBUG
 logging.level.com.zaxxer.hikari=WARN
 logging.level.stirling.software.SPDF.service.PdfJsonConversionService=INFO
 logging.level.stirling.software.common.service.JobExecutorService=INFO
@ -52,14 +51,43 @@ server.servlet.session.timeout:30m
 springdoc.api-docs.path=/v1/api-docs
 # Set the URL of the OpenAPI JSON for the Swagger UI
 springdoc.swagger-ui.url=/v1/api-docs
-springdoc.swagger-ui.path=/swagger-ui.html
+
+# Spring AI OpenAI Configuration
+# Uses GPT-5-nano as primary model and GPT-5-mini as fallback (configured in settings.yml)
+spring.ai.openai.enabled=true
+spring.ai.openai.api-key=# todo <API-KEY-HERE>
+spring.ai.openai.base-url=https://api.openai.com
+spring.ai.openai.chat.enabled=true
+spring.ai.openai.chat.options.model=gpt-5-nano
+# Note: Some models only support default temperature value of 1.0
+spring.ai.openai.chat.options.temperature=1.0
+# For newer models, use max-completion-tokens instead of max-tokens
+spring.ai.openai.chat.options.max-completion-tokens=4000
+spring.ai.openai.embedding.enabled=true
+spring.ai.openai.embedding.options.model=text-embedding-ada-002
+# Increase timeout for OpenAI API calls (default is 10 seconds)
+spring.ai.openai.chat.options.connection-timeout=60s
+spring.ai.openai.chat.options.read-timeout=60s
+spring.ai.openai.embedding.options.connection-timeout=60s
+spring.ai.openai.embedding.options.read-timeout=60s
+
+# Spring AI Ollama Configuration (disabled to avoid bean conflicts)
+spring.ai.ollama.enabled=false
+spring.ai.ollama.base-url=http://localhost:11434
+spring.ai.ollama.chat.enabled=false
+spring.ai.ollama.chat.options.model=llama3
+spring.ai.ollama.chat.options.temperature=1.0
+spring.ai.ollama.embedding.enabled=false
+spring.ai.ollama.embedding.options.model=nomic-embed-text
+
 # Force OpenAPI 3.0 specification version
+springdoc.swagger-ui.path=/swagger-ui.html
 springdoc.api-docs.version=OPENAPI_3_0
+
 posthog.api.key=phc_fiR65u5j6qmXTYL56MNrLZSWqLaDW74OrZH0Insd2xq
 posthog.host=https://eu.i.posthog.com

 spring.main.allow-bean-definition-overriding=true
-spring.ai.openai.enabled=false

 # Set up a consistent temporary directory location
 java.io.tmpdir=${stirling.tempfiles.directory:${java.io.tmpdir}/stirling-pdf}
--- a/app/proprietary/build.gradle
+++ b/app/proprietary/build.gradle
@ -52,8 +52,9 @@ dependencies {
    api 'org.springframework.boot:spring-boot-starter-cache'
    api 'com.github.ben-manes.caffeine:caffeine'
    api 'io.swagger.core.v3:swagger-core-jakarta:2.2.38'
-    implementation 'org.springframework.ai:spring-ai-openai'
-    implementation 'org.springframework.ai:spring-ai-ollama'
+    implementation 'org.springframework.ai:spring-ai-starter-model-openai'
+    implementation 'org.springframework.ai:spring-ai-starter-model-ollama'
+    implementation 'org.springframework.ai:spring-ai-redis-store'
    implementation 'com.bucket4j:bucket4j_jdk17-core:8.15.0'

    // https://mvnrepository.com/artifact/com.bucket4j/bucket4j_jdk17
--- a/app/proprietary/src/main/java/stirling/software/proprietary/config/SpringAIConfig.java
+++ b/app/proprietary/src/main/java/stirling/software/proprietary/config/SpringAIConfig.java
@ -0,0 +1,82 @@
+package stirling.software.proprietary.config;
+
+import java.time.Duration;
+
+import org.springframework.ai.chat.model.ChatModel;
+import org.springframework.ai.embedding.EmbeddingModel;
+import org.springframework.beans.factory.annotation.Qualifier;
+import org.springframework.boot.web.client.RestTemplateBuilder;
+import org.springframework.context.annotation.Bean;
+import org.springframework.context.annotation.Configuration;
+import org.springframework.context.annotation.Primary;
+import org.springframework.http.client.SimpleClientHttpRequestFactory;
+import org.springframework.web.client.RestClient;
+import org.springframework.web.client.RestTemplate;
+
+import lombok.extern.slf4j.Slf4j;
+
+/**
+ * Spring AI Configuration for Stirling PDF Chatbot
+ *
+ * <p>This configuration enables Spring AI auto-configuration for chatbot features. The actual
+ * ChatModel and EmbeddingModel beans are provided by Spring Boot's auto-configuration based on the
+ * spring.ai.* properties in application-proprietary.properties
+ *
+ * <p>For OpenAI: - spring.ai.openai.enabled=true - spring.ai.openai.api-key=your-api-key
+ *
+ * <p>For Ollama (as fallback): - spring.ai.ollama.enabled=true -
+ * spring.ai.ollama.base-url=http://localhost:11434
+ */
+@Configuration
+@Slf4j
+public class SpringAIConfig {
+
+    public SpringAIConfig() {
+        log.info("Spring AI Configuration enabled for Stirling PDF Chatbot");
+        log.info(
+                "ChatModel and EmbeddingModel beans will be auto-configured based on spring.ai.* properties");
+    }
+
+    /** Primary ChatModel bean that delegates to OpenAI's auto-configured bean */
+    @Bean
+    @Primary
+    public ChatModel primaryChatModel(@Qualifier("openAiChatModel") ChatModel openAiChatModel) {
+        log.info("Using OpenAI ChatModel as primary");
+        return openAiChatModel;
+    }
+
+    /** Primary EmbeddingModel bean that delegates to OpenAI's auto-configured bean */
+    @Bean
+    @Primary
+    public EmbeddingModel primaryEmbeddingModel(
+            @Qualifier("openAiEmbeddingModel") EmbeddingModel openAiEmbeddingModel) {
+        log.info("Using OpenAI EmbeddingModel as primary");
+        return openAiEmbeddingModel;
+    }
+
+    /**
+     * Custom RestTemplate for Spring AI OpenAI client with increased timeouts. This helps prevent
+     * timeout errors when processing large documents or complex queries.
+     */
+    @Bean(name = "openAiRestTemplate")
+    public RestTemplate openAiRestTemplate(RestTemplateBuilder builder) {
+        log.info("Creating custom RestTemplate for OpenAI with 60s timeouts");
+        return builder.connectTimeout(Duration.ofSeconds(60))
+                .readTimeout(Duration.ofSeconds(60))
+                .build();
+    }
+
+    /**
+     * Custom RestClient for Spring AI OpenAI with increased timeouts. Spring AI 1.0.3+ prefers
+     * RestClient over RestTemplate.
+     */
+    @Bean(name = "openAiRestClient")
+    public RestClient openAiRestClient() {
+        log.info("Creating custom RestClient for OpenAI with 60s timeouts");
+        SimpleClientHttpRequestFactory factory = new SimpleClientHttpRequestFactory();
+        factory.setConnectTimeout(Duration.ofSeconds(60));
+        factory.setReadTimeout(Duration.ofSeconds(60));
+
+        return RestClient.builder().requestFactory(factory).build();
+    }
+}
--- a/app/proprietary/src/main/java/stirling/software/proprietary/controller/ChatbotController.java
+++ b/app/proprietary/src/main/java/stirling/software/proprietary/controller/ChatbotController.java
@ -3,8 +3,6 @@ package stirling.software.proprietary.controller;
 import java.util.ArrayList;
 import java.util.List;

-import org.springframework.boot.autoconfigure.condition.ConditionalOnBean;
-import org.springframework.boot.autoconfigure.condition.ConditionalOnProperty;
 import org.springframework.http.HttpStatus;
 import org.springframework.http.ResponseEntity;
 import org.springframework.web.bind.annotation.DeleteMapping;
@ -31,11 +29,11 @@ import stirling.software.proprietary.service.chatbot.ChatbotSessionRegistry;
 import stirling.software.proprietary.service.chatbot.exception.ChatbotException;

@RestController
-@RequestMapping("/api/internal/chatbot")
+@RequestMapping("/api/v1/internal/chatbot")
@RequiredArgsConstructor
@Slf4j
-@ConditionalOnProperty(value = "premium.proFeatures.chatbot.enabled", havingValue = "true")
-@ConditionalOnBean(ChatbotService.class)
+// @ConditionalOnProperty(value = "premium.proFeatures.chatbot.enabled", havingValue = "true")
+// @ConditionalOnBean(ChatbotService.class)
 public class ChatbotController {

    private final ChatbotService chatbotService;
--- a/app/proprietary/src/main/java/stirling/software/proprietary/controller/ChatbotExceptionHandler.java
+++ b/app/proprietary/src/main/java/stirling/software/proprietary/controller/ChatbotExceptionHandler.java
@ -4,7 +4,6 @@ import java.time.Instant;
 import java.util.Map;

 import org.springframework.boot.autoconfigure.condition.ConditionalOnBean;
-import org.springframework.boot.autoconfigure.condition.ConditionalOnProperty;
 import org.springframework.http.HttpStatus;
 import org.springframework.http.ResponseEntity;
 import org.springframework.web.bind.annotation.ExceptionHandler;
@ -18,7 +17,7 @@ import stirling.software.proprietary.service.chatbot.exception.NoTextDetectedExc

@RestControllerAdvice(assignableTypes = ChatbotController.class)
@Slf4j
-@ConditionalOnProperty(value = "premium.proFeatures.chatbot.enabled", havingValue = "true")
+// @ConditionalOnProperty(value = "premium.proFeatures.chatbot.enabled", havingValue = "true")
@ConditionalOnBean(ChatbotService.class)
 public class ChatbotExceptionHandler {

--- a/app/proprietary/src/main/java/stirling/software/proprietary/service/chatbot/ChatbotCacheService.java
+++ b/app/proprietary/src/main/java/stirling/software/proprietary/service/chatbot/ChatbotCacheService.java
@ -9,7 +9,6 @@ import java.util.Optional;
 import java.util.UUID;
 import java.util.concurrent.ConcurrentHashMap;

-import org.springframework.boot.autoconfigure.condition.ConditionalOnProperty;
 import org.springframework.stereotype.Service;

 import com.github.benmanes.caffeine.cache.Cache;
@ -26,7 +25,7 @@ import stirling.software.proprietary.model.chatbot.ChatbotTextChunk;
 import stirling.software.proprietary.service.chatbot.exception.ChatbotException;

@Service
-@ConditionalOnProperty(value = "premium.proFeatures.chatbot.enabled", havingValue = "true")
+// @ConditionalOnProperty(value = "premium.proFeatures.chatbot.enabled", havingValue = "true")
@Slf4j
 public class ChatbotCacheService {

--- a/app/proprietary/src/main/java/stirling/software/proprietary/service/chatbot/ChatbotConversationService.java
+++ b/app/proprietary/src/main/java/stirling/software/proprietary/service/chatbot/ChatbotConversationService.java
@ -1,5 +1,7 @@
 package stirling.software.proprietary.service.chatbot;

+import static stirling.software.proprietary.service.chatbot.ChatbotFeatureProperties.ChatbotSettings.ModelProvider.OLLAMA;
+
 import java.io.IOException;
 import java.time.Instant;
 import java.util.ArrayList;
@ -15,11 +17,8 @@ import org.springframework.ai.chat.model.ChatModel;
 import org.springframework.ai.chat.model.ChatResponse;
 import org.springframework.ai.chat.prompt.Prompt;
 import org.springframework.ai.ollama.OllamaChatModel;
-import org.springframework.ai.ollama.api.OllamaOptions;
 import org.springframework.ai.openai.OpenAiChatModel;
 import org.springframework.ai.openai.OpenAiChatOptions;
-import org.springframework.boot.autoconfigure.condition.ConditionalOnBean;
-import org.springframework.boot.autoconfigure.condition.ConditionalOnProperty;
 import org.springframework.stereotype.Service;
 import org.springframework.util.StringUtils;

@ -35,14 +34,11 @@ import stirling.software.proprietary.model.chatbot.ChatbotResponse;
 import stirling.software.proprietary.model.chatbot.ChatbotSession;
 import stirling.software.proprietary.model.chatbot.ChatbotTextChunk;
 import stirling.software.proprietary.service.chatbot.ChatbotFeatureProperties.ChatbotSettings;
-import stirling.software.proprietary.service.chatbot.ChatbotFeatureProperties.ChatbotSettings.ModelProvider;
 import stirling.software.proprietary.service.chatbot.exception.ChatbotException;

@Service
@Slf4j
@RequiredArgsConstructor
-@ConditionalOnProperty(value = "premium.proFeatures.chatbot.enabled", havingValue = "true")
-@ConditionalOnBean(ChatModel.class)
 public class ChatbotConversationService {

    private final ChatModel chatModel;
@ -159,7 +155,7 @@ public class ChatbotConversationService {
    }

    private void ensureModelSwitchCapability(ChatbotSettings settings) {
-        ModelProvider provider = settings.models().provider();
+        ChatbotSettings.ModelProvider provider = settings.models().provider();
        switch (provider) {
            case OPENAI -> {
                if (!(chatModel instanceof OpenAiChatModel)) {
@ -262,22 +258,15 @@ public class ChatbotConversationService {
                        + "Question: "
                        + question;

-        Object options = buildChatOptions(settings, model);
+        OpenAiChatOptions options = buildChatOptions(model);

        return new Prompt(
                List.of(new SystemMessage(systemPrompt), new UserMessage(userPrompt)), options);
    }

-    private Object buildChatOptions(ChatbotSettings settings, String model) {
-        return switch (settings.models().provider()) {
-            case OPENAI ->
-                    OpenAiChatOptions.builder()
-                            .model(model)
-                            .temperature(0.2)
-                            .responseFormat("json_object")
-                            .build();
-            case OLLAMA -> OllamaOptions.builder().model(model).temperature(0.2).build();
-        };
+    private OpenAiChatOptions buildChatOptions(String model) {
+        // Note: Some models only support default temperature value of 1.0
+        return OpenAiChatOptions.builder().model(model).temperature(1.0).build();
    }

    private ModelReply parseModelResponse(String raw) {
--- a/app/proprietary/src/main/java/stirling/software/proprietary/service/chatbot/ChatbotIngestionService.java
+++ b/app/proprietary/src/main/java/stirling/software/proprietary/service/chatbot/ChatbotIngestionService.java
@ -9,8 +9,6 @@ import java.util.UUID;

 import org.springframework.ai.embedding.EmbeddingModel;
 import org.springframework.ai.embedding.EmbeddingResponse;
-import org.springframework.boot.autoconfigure.condition.ConditionalOnBean;
-import org.springframework.boot.autoconfigure.condition.ConditionalOnProperty;
 import org.springframework.stereotype.Service;
 import org.springframework.util.StringUtils;

@ -25,8 +23,6 @@ import stirling.software.proprietary.service.chatbot.exception.ChatbotException;
 import stirling.software.proprietary.service.chatbot.exception.NoTextDetectedException;

@Service
-@ConditionalOnProperty(value = "premium.proFeatures.chatbot.enabled", havingValue = "true")
-@ConditionalOnBean(EmbeddingModel.class)
@Slf4j
@RequiredArgsConstructor
 public class ChatbotIngestionService {
--- a/app/proprietary/src/main/java/stirling/software/proprietary/service/chatbot/ChatbotRetrievalService.java
+++ b/app/proprietary/src/main/java/stirling/software/proprietary/service/chatbot/ChatbotRetrievalService.java
@ -7,8 +7,6 @@ import java.util.Optional;

 import org.springframework.ai.embedding.EmbeddingModel;
 import org.springframework.ai.embedding.EmbeddingResponse;
-import org.springframework.boot.autoconfigure.condition.ConditionalOnBean;
-import org.springframework.boot.autoconfigure.condition.ConditionalOnProperty;
 import org.springframework.stereotype.Service;
 import org.springframework.util.CollectionUtils;

@ -23,8 +21,6 @@ import stirling.software.proprietary.service.chatbot.exception.ChatbotException;
@Service
@RequiredArgsConstructor
@Slf4j
-@ConditionalOnProperty(value = "premium.proFeatures.chatbot.enabled", havingValue = "true")
-@ConditionalOnBean(EmbeddingModel.class)
 public class ChatbotRetrievalService {

    private final ChatbotCacheService cacheService;
--- a/app/proprietary/src/main/java/stirling/software/proprietary/service/chatbot/ChatbotService.java
+++ b/app/proprietary/src/main/java/stirling/software/proprietary/service/chatbot/ChatbotService.java
@ -3,8 +3,6 @@ package stirling.software.proprietary.service.chatbot;
 import java.util.HashMap;
 import java.util.Map;

-import org.springframework.boot.autoconfigure.condition.ConditionalOnBean;
-import org.springframework.boot.autoconfigure.condition.ConditionalOnProperty;
 import org.springframework.stereotype.Service;

 import lombok.RequiredArgsConstructor;
@ -18,8 +16,6 @@ import stirling.software.proprietary.service.AuditService;
 import stirling.software.proprietary.service.chatbot.exception.ChatbotException;

@Service
-@ConditionalOnProperty(value = "premium.proFeatures.chatbot.enabled", havingValue = "true")
-@ConditionalOnBean({ChatbotIngestionService.class, ChatbotConversationService.class})
@Slf4j
@RequiredArgsConstructor
 public class ChatbotService {
--- a/app/proprietary/src/main/resources/application-proprietary.properties
+++ b/app/proprietary/src/main/resources/application-proprietary.properties
@ -1,2 +0,0 @@
-spring.ai.openai.enabled=true
-spring.ai.openai.api-key=your-proprietary-api-key-here
--- a/app/proprietary/src/test/java/stirling/software/proprietary/service/chatbot/ChatbotServiceTest.java
+++ b/app/proprietary/src/test/java/stirling/software/proprietary/service/chatbot/ChatbotServiceTest.java
@ -48,7 +48,11 @@ class ChatbotServiceTest {
                        true,
                        4000,
                        0.5D,
-                        new ChatbotSettings.ModelSettings("gpt-5-nano", "gpt-5-mini", "embed"),
+                        new ChatbotSettings.ModelSettings(
+                                ChatbotSettings.ModelProvider.OPENAI,
+                                "gpt-5-nano",
+                                "gpt-5-mini",
+                                "embed"),
                        new ChatbotSettings.RagSettings(512, 128, 4),
                        new ChatbotSettings.CacheSettings(60, 10, 1000),
                        new ChatbotSettings.OcrSettings(false),
@ -60,7 +64,11 @@ class ChatbotServiceTest {
                        true,
                        4000,
                        0.5D,
-                        new ChatbotSettings.ModelSettings("gpt-5-nano", "gpt-5-mini", "embed"),
+                        new ChatbotSettings.ModelSettings(
+                                ChatbotSettings.ModelProvider.OPENAI,
+                                "gpt-5-nano",
+                                "gpt-5-mini",
+                                "embed"),
                        new ChatbotSettings.RagSettings(512, 128, 4),
                        new ChatbotSettings.CacheSettings(60, 10, 1000),
                        new ChatbotSettings.OcrSettings(false),
--- a/frontend/src/core/services/chatbotService.ts
+++ b/frontend/src/core/services/chatbotService.ts
@ -40,12 +40,12 @@ export interface ChatbotMessageResponse {
 }

 export async function createChatbotSession(payload: ChatbotSessionPayload) {
-  const { data } = await apiClient.post<ChatbotSessionInfo>('/api/internal/chatbot/session', payload);
+  const { data } = await apiClient.post<ChatbotSessionInfo>('/api/v1/internal/chatbot/session', payload);
  return data;
 }

 export async function sendChatbotPrompt(payload: ChatbotQueryPayload) {
-  const { data } = await apiClient.post<ChatbotMessageResponse>('/api/internal/chatbot/query', payload);
+  const { data } = await apiClient.post<ChatbotMessageResponse>('/api/v1/internal/chatbot/query', payload);
  return data;
 }