fix:提升版本，修复脚本

scripts/analyze_group_similarity.py

@@ -60,9 +60,9 @@ def load_group_data(group_dir):
 
     for item in data:
         count = item["count"]
-        situations.extend([item["situation"]] * count)
-        styles.extend([item["style"]] * count)
-        combined.extend([f"{item['situation']} {item['style']}"] * count)
+        situations.extend([item["situation"]] * int(count))
+        styles.extend([item["style"]] * int(count))
+        combined.extend([f"{item['situation']} {item['style']}"] * int(count))
 
     return situations, styles, combined, total_count
 

src/chat/focus_chat/expressors/exprssion_learner.py

@@ -280,6 +280,21 @@ class ExpressionLearner:
                     new_expr["last_active_time"] = current_time
                     old_data.append(new_expr)
             
+            # 处理超限问题
+            if len(old_data) > MAX_EXPRESSION_COUNT:
+                # 计算每个表达方式的权重（count的倒数，这样count越小的越容易被选中）
+                weights = [1 / (expr.get("count", 1) + 0.1) for expr in old_data]
+                # 归一化权重
+                total_weight = sum(weights)
+                weights = [w / total_weight for w in weights]
+                
+                # 随机选择要移除的表达方式
+                remove_count = len(old_data) - MAX_EXPRESSION_COUNT
+                remove_indices = random.choices(range(len(old_data)), weights=weights, k=remove_count)
+                # 从后往前删除，避免索引变化
+                for idx in sorted(remove_indices, reverse=True):
+                    old_data.pop(idx)
+            
             with open(file_path, "w", encoding="utf-8") as f:
                 json.dump(old_data, f, ensure_ascii=False, indent=2)
         

src/chat/heart_flow/observation/chatting_observation.py

@@ -132,7 +132,10 @@ class ChattingObservation(Observation):
                 # logger.debug(f"找到的锚定消息：find_msg: {find_msg}")
                 break
             else:
-                similarity = difflib.SequenceMatcher(None, text, message["raw_message"]).ratio()
+                if message["raw_message"]:
+                    similarity = difflib.SequenceMatcher(None, text, message["raw_message"]).ratio()
+                else:
+                    similarity = difflib.SequenceMatcher(None, text, message["processed_plain_text"]).ratio()
                 msg_list.append({"message": message, "similarity": similarity})
             # logger.debug(f"对锚定消息检查：message: {message['processed_plain_text']},similarity: {similarity}")
         if not find_msg:

src/config/config.py

@@ -47,7 +47,7 @@ TEMPLATE_DIR = "template"
 
 # 考虑到，实际上配置文件中的mai_version是不会自动更新的,所以采用硬编码
 # 对该字段的更新，请严格参照语义化版本规范：https://semver.org/lang/zh-CN/
-MMC_VERSION = "0.7.2-snapshot.1"
+MMC_VERSION = "0.7.3-snapshot.1"
 
 
 def update_config():