feat: add comprehensive error handling to LoraInfo_UTK node - 增加网络连接失败、文件读取错误等异常处理，确保程序稳定运行

2025-07-15 18:49:43 +08:00
parent c701d628cf
commit 5a8ab327fd
2 changed files with 200 additions and 108 deletions
@@ -41,11 +41,25 @@ def save_dict_to_json(data_dict, file_path):

 def get_model_version_info(hash_value):
    api_url = f"https://civitai.com/api/v1/model-versions/by-hash/{hash_value}"
-    response = requests.get(api_url)
+    try:
+        response = requests.get(api_url, timeout=10)  # 设置10秒超时
        
-    if response.status_code == 200:
-        return response.json()
-    else:
+        if response.status_code == 200:
+            return response.json()
+        else:
+            print(f"[LoraInfo_UTK] CivitAI API返回错误状态码: {response.status_code}")
+            return {}
+    except requests.exceptions.ConnectionError:
+        print("[LoraInfo_UTK] 无法连接到CivitAI服务器，请检查网络连接")
+        return {}
+    except requests.exceptions.Timeout:
+        print("[LoraInfo_UTK] 连接CivitAI服务器超时，请稍后重试")
+        return {}
+    except requests.exceptions.RequestException as e:
+        print(f"[LoraInfo_UTK] 请求CivitAI API时发生错误: {e}")
+        return {}
+    except Exception as e:
+        print(f"[LoraInfo_UTK] 获取模型信息时发生未知错误: {e}")
        return {}

 def calculate_sha256(file_path):
@@ -98,90 +112,135 @@ def sort_tags_by_frequency(meta_tags):
        return []

 def get_lora_info(lora_name):
-    db = load_json_from_file(db_path)
-    output = None
-    examplePrompt = None
-    trainedWords = None
-    baseModel = None
-    metaInfo = None
-
-    loraInfo = db.get(lora_name, {})
-
-    if isinstance(loraInfo, str):
-        loraInfo = {}
-    
-    output = loraInfo.get('output', None)
-    examplePrompt = loraInfo.get('examplePrompt', None)
-    trainedWords = loraInfo.get('trainedWords', None)
-    baseModel = loraInfo.get('baseModel', None)
-    metaInfo = loraInfo.get('metaInfo', None)
-
-    if output is None or baseModel is None:
-        output = ""
-        lora_path = folder_paths.get_full_path("loras", lora_name)
-        LORAsha256 = calculate_sha256(lora_path)
-        model_info = get_model_version_info(LORAsha256)
-        if model_info.get("trainedWords", None) is None:
-            trainedWords = ""
-        else:
-            trainedWords = ",".join(model_info.get("trainedWords"))
-        baseModel = model_info.get("baseModel", "")
-        images = model_info.get('images')
+    try:
+        db = load_json_from_file(db_path)
+        output = None
        examplePrompt = None
-        modelID = model_info.get("modelId")
+        trainedWords = None
+        baseModel = None
+        metaInfo = None

-        if modelID:
-            output += f"URL: https://civitai.com/models/{modelID}\n"
-        if trainedWords:
-            output += "Triggers: " + trainedWords
-            output += "\n"
+        loraInfo = db.get(lora_name, {})

-        if baseModel:
-            output += f"Base Model: {baseModel}\n"
-        if images:
-            output += "\nExamples:\n"
-            for image in images:
-                output += f"\nOutput: {image.get('url')}\n"
-                meta = image.get("meta")
-                if meta:
-                    for key, value in meta.items():
-                        if examplePrompt is None and key == "prompt":
-                            examplePrompt = value
-                        output += f"{key}: {value}\n"
-                output += '\n'
+        if isinstance(loraInfo, str):
+            loraInfo = {}
        
-        # 获取元数据信息
-        metadata = get_metadata(lora_name)
-        if metadata:
-            metaInfo = json.dumps(metadata, indent=2, ensure_ascii=False)
-        else:
-            metaInfo = ""
+        output = loraInfo.get('output', None)
+        examplePrompt = loraInfo.get('examplePrompt', None)
+        trainedWords = loraInfo.get('trainedWords', None)
+        baseModel = loraInfo.get('baseModel', None)
+        metaInfo = loraInfo.get('metaInfo', None)

-        db[lora_name] = {
-            "output": output,
-            "trainedWords": trainedWords,
-            "examplePrompt": examplePrompt,
-            "baseModel": baseModel,
-            "metaInfo": metaInfo
-            }
-        save_dict_to_json(db, db_path)
+        if output is None or baseModel is None:
+            output = ""
+            try:
+                lora_path = folder_paths.get_full_path("loras", lora_name)
+                if not lora_path:
+                    print(f"[LoraInfo_UTK] 无法找到LoRA文件: {lora_name}")
+                    return ("", "", "", "", "")
                
-    return (output, trainedWords, examplePrompt, baseModel, metaInfo)
+                LORAsha256 = calculate_sha256(lora_path)
+                model_info = get_model_version_info(LORAsha256)
+                
+                if model_info.get("trainedWords", None) is None:
+                    trainedWords = ""
+                else:
+                    trainedWords = ",".join(model_info.get("trainedWords"))
+                baseModel = model_info.get("baseModel", "")
+                images = model_info.get('images')
+                examplePrompt = None
+                modelID = model_info.get("modelId")
+                
+                if modelID:
+                    output += f"URL: https://civitai.com/models/{modelID}\n"
+                if trainedWords:
+                    output += "Triggers: " + trainedWords
+                    output += "\n"
+                
+                if baseModel:
+                    output += f"Base Model: {baseModel}\n"
+                if images:
+                    output += "\nExamples:\n"
+                    for image in images:
+                        output += f"\nOutput: {image.get('url')}\n"
+                        meta = image.get("meta")
+                        if meta:
+                            for key, value in meta.items():
+                                if examplePrompt is None and key == "prompt":
+                                    examplePrompt = value
+                                output += f"{key}: {value}\n"
+                        output += '\n'
+
+                # 获取元数据信息
+                try:
+                    metadata = get_metadata(lora_name)
+                    if metadata:
+                        metaInfo = json.dumps(metadata, indent=2, ensure_ascii=False)
+                    else:
+                        metaInfo = ""
+                except Exception as e:
+                    print(f"[LoraInfo_UTK] 读取元数据时发生错误: {e}")
+                    metaInfo = ""
+
+                db[lora_name] = {
+                    "output": output,
+                    "trainedWords": trainedWords,
+                    "examplePrompt": examplePrompt,
+                    "baseModel": baseModel,
+                    "metaInfo": metaInfo
+                    }
+                save_dict_to_json(db, db_path)
+                
+            except Exception as e:
+                print(f"[LoraInfo_UTK] 处理LoRA文件时发生错误: {e}")
+                output = f"处理LoRA文件时发生错误: {e}"
+                trainedWords = ""
+                examplePrompt = ""
+                baseModel = ""
+                metaInfo = ""
+        
+        return (output, trainedWords, examplePrompt, baseModel, metaInfo)
+        
+    except Exception as e:
+        print(f"[LoraInfo_UTK] 获取LoRA信息时发生严重错误: {e}")
+        return ("", "", "", "", "")


@server.PromptServer.instance.routes.post('/lora_info_utk')
 async def fetch_lora_info(request):
-    post = await request.post()
-    lora_name = post.get("lora_name")
-    (output, triggerWords, examplePrompt, baseModel, metaInfo) = get_lora_info(lora_name)
+    try:
+        post = await request.post()
+        lora_name = post.get("lora_name")
        
-    return web.json_response({
-        "output": output, 
-        "triggerWords": triggerWords, 
-        "examplePrompt": examplePrompt, 
-        "baseModel": baseModel,
-        "metaInfo": metaInfo
-    })
+        if not lora_name:
+            return web.json_response({
+                "error": "未提供LoRA名称",
+                "output": "",
+                "triggerWords": "",
+                "examplePrompt": "",
+                "baseModel": "",
+                "metaInfo": ""
+            })
+        
+        (output, triggerWords, examplePrompt, baseModel, metaInfo) = get_lora_info(lora_name)
+
+        return web.json_response({
+            "output": output, 
+            "triggerWords": triggerWords, 
+            "examplePrompt": examplePrompt, 
+            "baseModel": baseModel,
+            "metaInfo": metaInfo
+        })
+    except Exception as e:
+        print(f"[LoraInfo_UTK] Web API调用时发生错误: {e}")
+        return web.json_response({
+            "error": f"处理请求时发生错误: {e}",
+            "output": "",
+            "triggerWords": "",
+            "examplePrompt": "",
+            "baseModel": "",
+            "metaInfo": ""
+        })

 class LoraInfo_UTK:
    """
@@ -212,26 +271,37 @@ class LoraInfo_UTK:
    CATEGORY = "UniversalToolkit/Tools"

    def lora_info(self, lora_name):
-        (output, triggerWords, examplePrompt, baseModel, metaInfo) = get_lora_info(lora_name)
+        try:
+            (output, triggerWords, examplePrompt, baseModel, metaInfo) = get_lora_info(lora_name)
            
-        # 构建信息文本
-        info_text = f"LoRA: {lora_name}\n"
-        if baseModel:
-            info_text += f"Base Model: {baseModel}\n"
-        if triggerWords:
-            info_text += f"Trigger Words: {triggerWords}\n"
-        if examplePrompt:
-            info_text += f"Example Prompt: {examplePrompt}\n"
-        if output:
-            info_text += f"\n详细信息:\n{output}"
+            # 构建信息文本
+            info_text = f"LoRA: {lora_name}\n"
+            if baseModel:
+                info_text += f"Base Model: {baseModel}\n"
+            if triggerWords:
+                info_text += f"Trigger Words: {triggerWords}\n"
+            if examplePrompt:
+                info_text += f"Example Prompt: {examplePrompt}\n"
+            if output:
+                info_text += f"\n详细信息:\n{output}"
            
-        return {
-            "ui": {
-                "text": (info_text,), 
-                "model": (baseModel,)
-            }, 
-            "result": (lora_name, triggerWords or "", examplePrompt or "", info_text, metaInfo or "")
-        }
+            return {
+                "ui": {
+                    "text": (info_text,), 
+                    "model": (baseModel,)
+                }, 
+                "result": (lora_name, triggerWords or "", examplePrompt or "", info_text, metaInfo or "")
+            }
+        except Exception as e:
+            print(f"[LoraInfo_UTK] 节点执行时发生错误: {e}")
+            error_text = f"LoRA: {lora_name}\n获取信息时发生错误: {e}"
+            return {
+                "ui": {
+                    "text": (error_text,), 
+                    "model": ("",)
+                }, 
+                "result": (lora_name, "", "", error_text, "")
+            }


 # Node mappings
@@ -36,9 +36,24 @@ app.registerExtension({
            .fetchApi("/lora_info_utk", { method: "POST", body })
            .then((response) => response.json())
            .then((resp) => {
-              baseModelWidget.value = resp.baseModel;
-              outputWidget.value = resp.output;
-              metaInfoWidget.value = resp.metaInfo;
+              if (resp.error) {
+                // 显示错误信息
+                baseModelWidget.value = "错误";
+                outputWidget.value = `获取信息失败: ${resp.error}`;
+                metaInfoWidget.value = "无法获取元数据";
+              } else {
+                // 正常显示信息
+                baseModelWidget.value = resp.baseModel;
+                outputWidget.value = resp.output;
+                metaInfoWidget.value = resp.metaInfo;
+              }
+            })
+            .catch((error) => {
+              // 处理网络错误
+              console.error("[LoraInfo_UTK] API调用失败:", error);
+              baseModelWidget.value = "网络错误";
+              outputWidget.value = "无法连接到服务器，请检查网络连接";
+              metaInfoWidget.value = "无法获取元数据";
            });
        };
      }
@@ -46,9 +61,16 @@ app.registerExtension({
      const onExecuted = nodeType.prototype.onExecuted;
      nodeType.prototype.onExecuted = function (message) {
        onExecuted?.apply(this, [message]);
-        this.showValueWidget.value = message.text[0];
-        this.baseModelWidget.value = message.model[0];
-        this.metaInfoWidget.value = message.metaInfo ? message.metaInfo[0] : "";
+        try {
+          this.showValueWidget.value = message.text[0];
+          this.baseModelWidget.value = message.model[0];
+          this.metaInfoWidget.value = message.metaInfo ? message.metaInfo[0] : "";
+        } catch (error) {
+          console.error("[LoraInfo_UTK] 更新界面时发生错误:", error);
+          this.showValueWidget.value = "界面更新失败";
+          this.baseModelWidget.value = "";
+          this.metaInfoWidget.value = "";
+        }
      }
    }
  },