Feat: AI 意图识别节点多路选择初步实现

liteflow-ai/liteflow-ai-core/src/main/java/com/yomahub/liteflow/ai/proxy/handler/AbstractAIComponentHandler.java

@@ -172,6 +172,7 @@ public abstract class AbstractAIComponentHandler<T extends Annotation> {
     /**
      * 获取拦截方法名称
      *
+     * @param wrapBean 包装bean，现在暂时不用，后续如果单节点需要根据条件判断拦截方法，可以用上
      * @return 拦截方法名称
      */
     protected abstract ElementMatcher<? super MethodDescription> getInterceptMethodName(AIProxyWrapBean<T> wrapBean);

liteflow-ai/liteflow-ai-core/src/main/java/com/yomahub/liteflow/ai/proxy/handler/ClassifyComponentHandler.java

@@ -1,7 +1,7 @@
 package com.yomahub.liteflow.ai.proxy.handler;
 
-import com.yomahub.liteflow.ai.annotation.model.node.AIClassify;
 import com.yomahub.liteflow.ai.annotation.AIComponent;
+import com.yomahub.liteflow.ai.annotation.model.node.AIClassify;
 import com.yomahub.liteflow.ai.domain.enums.AITypeEnum;
 import com.yomahub.liteflow.ai.proxy.invocation.ClassifyAIInvocationHandler;
 import com.yomahub.liteflow.ai.proxy.wrap.AIProxyWrapBean;
@@ -22,7 +22,6 @@ import java.lang.reflect.InvocationHandler;
 public class ClassifyComponentHandler extends AbstractAIComponentHandler<AIClassify> {
 
     private static final String INTERCEPT_SWITCH_METHOD_NAME = "processSwitch";
-    private static final String INTERCEPT_MULTI_SWITCH_METHOD_NAME = "processMultiSwitch";
 
     @Override
     public AITypeEnum getAIType() {
@@ -47,10 +46,6 @@ public class ClassifyComponentHandler extends AbstractAIComponentHandler<AIClass
 
     @Override
     protected ElementMatcher<? super MethodDescription> getInterceptMethodName(AIProxyWrapBean<AIClassify> wrapBean) {
-        if (wrapBean.getAnnotation().multiLabel()) {
-            return ElementMatchers.named(INTERCEPT_MULTI_SWITCH_METHOD_NAME);
-        } else {
-            return ElementMatchers.named(INTERCEPT_SWITCH_METHOD_NAME);
-        }
+        return ElementMatchers.named(INTERCEPT_SWITCH_METHOD_NAME);
     }
 }

liteflow-ai/liteflow-ai-core/src/main/java/com/yomahub/liteflow/ai/proxy/invocation/ClassifyAIInvocationHandler.java

@@ -9,6 +9,8 @@ import com.yomahub.liteflow.ai.parse.context.ProcessorContext;
 import com.yomahub.liteflow.ai.proxy.wrap.ClassifyProxyWrapBean;
 import com.yomahub.liteflow.ai.util.SetUtil;
 
+import java.util.List;
+
 /**
  * 分类组件的调用处理器
  *
@@ -33,6 +35,7 @@ public class ClassifyAIInvocationHandler extends AbstractAIInvocationHandler<Cla
         }
     }
 
+    @SuppressWarnings("unchecked")
     @Override
     protected Object doExecuteAIProcess(ProcessorContext<?> processorContext, Object[] args) {
         ChatModel chatModel = ModelFactory.getChatModel(wrapBean);
@@ -40,7 +43,8 @@ public class ClassifyAIInvocationHandler extends AbstractAIInvocationHandler<Cla
         ParsedClassifyAnnotationConfig annotationConfig = (ParsedClassifyAnnotationConfig) processorContext.getParsedAnnotationConfig();
         // 如果是多标签则返回结构化转换的list, 单标签返回 String
         if (annotationConfig.isMultiLabel()) {
-            return response.as(processorContext.getModelRequest().toChatRequest().getOutputParser());
+            List<String> resList = (List<String>) response.as(processorContext.getModelRequest().toChatRequest().getOutputParser());
+            return String.join(",", resList);
         } else {
             return response.getContent().getContent();
         }

liteflow-ai/liteflow-ai-model/liteflow-ai-openai/src/main/java/com/yomahub/liteflow/ai/model/openai/model/chat/OpenAIChatRequest.java

@@ -78,7 +78,7 @@ public class OpenAIChatRequest extends ChatRequest {
                     ObjectNode responseFormat = ObjectMapperHolder.createObjectNode();
                     responseFormat.put("type", "json_schema");
                     ObjectNode jsonSchema = ObjectMapperHolder.createObjectNode();
-                    jsonSchema.put("name", ((Class<?>) outputParser.getTargetType()).getSimpleName());
+                    jsonSchema.put("name", outputParser.getTargetType().getTypeName());
                     jsonSchema.set("schema", outputParser.getJsonSchema());
 
                     responseFormat.set("json_schema", jsonSchema);

liteflow-testcase-el/liteflow-testcase-el-ai/src/test/java/com/yomahub/liteflow/test/ai/core/classify/ClassifyTest.java

@@ -33,5 +33,14 @@ public class ClassifyTest {
     public void testClassify() {
         LiteflowResponse response = flowExecutor.execute2Resp("chain1", null, ChatContext.class);
         Assertions.assertTrue(response.isSuccess());
+        Assertions.assertEquals("a==>aiSwitch[aiSwitch]==>java", response.getExecuteStepStr());
     }
+
+    @Test
+    public void testMultiClassify() {
+        LiteflowResponse response = flowExecutor.execute2Resp("chain2", null, ChatContext.class);
+        Assertions.assertTrue(response.isSuccess());
+        Assertions.assertEquals("a==>aiMultiSwitch[aiMultiSwitch]==>java==>python", response.getExecuteStepStr());
+    }
+
 }

liteflow-testcase-el/liteflow-testcase-el-ai/src/test/java/com/yomahub/liteflow/test/ai/core/classify/cmp/AIClassifyCmp.java

@@ -15,8 +15,8 @@ import com.yomahub.liteflow.ai.util.TriState;
  */
 
 @AIComponent(
-        nodeId = "ai",
-        nodeName = "ai",
+        nodeId = "aiSwitch",
+        nodeName = "aiSwitch",
 //        provider = "ollama",
 //        apiUrl = "http://localhost:11434",
 //        model = "qwen3:32b",
@@ -28,9 +28,9 @@ import com.yomahub.liteflow.ai.util.TriState;
         connectTimeout = "10m"
 )
 @AIClassify(
-        systemPrompt = "你是一个意图识别高手，你需要识别用户的意图",
         userPrompt = "{{question}}",
-        categories = {"java", "python"})
+        categories = {"java", "python"}
+)
 @AIInput(
         mapping = {
                 @InputField(name = "question", expression = "test", defaultValue = "请帮我写一段Java代码"),

liteflow-testcase-el/liteflow-testcase-el-ai/src/test/java/com/yomahub/liteflow/test/ai/core/classify/cmp/AIMultiClassifyCmp.java

@@ -0,0 +1,46 @@
+package com.yomahub.liteflow.test.ai.core.classify.cmp;
+
+import com.yomahub.liteflow.ai.annotation.AIComponent;
+import com.yomahub.liteflow.ai.annotation.model.io.AIInput;
+import com.yomahub.liteflow.ai.annotation.model.io.AIOutput;
+import com.yomahub.liteflow.ai.annotation.model.io.InputField;
+import com.yomahub.liteflow.ai.annotation.model.node.AIClassify;
+import com.yomahub.liteflow.ai.util.TriState;
+
+/**
+ * TODO
+ *
+ * @author 苍镜月
+ * @since TODO
+ */
+
+@AIComponent(
+        nodeId = "aiMultiSwitch",
+        nodeName = "aiMultiSwitch",
+//        provider = "ollama",
+//        apiUrl = "http://localhost:11434",
+//        model = "qwen3:32b",
+        provider = "openai",
+        apiUrl = "https://ark.cn-beijing.volces.com/api/v3",
+        model = "doubao-seed-1-6-250615",
+        enableThinking = TriState.FALSE,
+        readTimeout = "10m",
+        connectTimeout = "10m"
+)
+@AIClassify(
+        userPrompt = "{{question}}",
+        categories = {"java", "python"},
+        multiLabel = true
+)
+@AIInput(
+        mapping = {
+                @InputField(name = "question", expression = "test", defaultValue = "请帮我写一段Java代码, 同时给出 Python 代码"),
+        }
+)
+@AIOutput(
+        methodExpress = "setData",
+        useKeyIndex = true,
+        key = "result"
+)
+public interface AIMultiClassifyCmp {
+}

liteflow-testcase-el/liteflow-testcase-el-ai/src/test/resources/core/classify/flow.el.xml

@@ -2,6 +2,10 @@
 <!DOCTYPE flow PUBLIC  "liteflow" "liteflow.dtd">
 <flow>
     <chain name="chain1">
-        THEN(a, SWITCH(ai).TO(java, python));
+        THEN(a, SWITCH(aiSwitch).TO(java, python));
+    </chain>
+
+    <chain name="chain2">
+        THEN(a, SWITCH(aiMultiSwitch).TO(java, python));
     </chain>
 </flow>