feat(memory): 增强记忆类型解析,支持中英文映射
新增了一个 `_resolve_memory_type` 方法,以更健壮地处理从 LLM 返回的记忆类型字符串。此方法现在可以正确解析中文、英文(大小写不敏感)以及下划线格式的记忆类型。 - 增加了从中文到 `MemoryType` 枚举的直接映射。 - 实现了对多种英文格式(如 "personal_fact", "PERSONAL_FACT", "Personal Fact")的兼容解析。 - 当无法识别任何有效类型时,会记录警告并安全地回退到默认的 `CONTEXTUAL` 类型,提高了系统的容错性。
This commit is contained in:
@@ -32,7 +32,10 @@ import time
|
||||
from dataclasses import dataclass
|
||||
from datetime import datetime
|
||||
from enum import Enum
|
||||
from typing import Any
|
||||
from typing import Any, Type, TypeVar
|
||||
|
||||
E = TypeVar("E", bound=Enum)
|
||||
|
||||
|
||||
import orjson
|
||||
|
||||
@@ -49,6 +52,21 @@ from src.llm_models.utils_model import LLMRequest
|
||||
logger = get_logger(__name__)
|
||||
|
||||
|
||||
CHINESE_TO_MEMORY_TYPE: dict[str, MemoryType] = {
|
||||
"个人事实": MemoryType.PERSONAL_FACT,
|
||||
"事件": MemoryType.EVENT,
|
||||
"偏好": MemoryType.PREFERENCE,
|
||||
"观点": MemoryType.OPINION,
|
||||
"关系": MemoryType.RELATIONSHIP,
|
||||
"情感": MemoryType.EMOTION,
|
||||
"知识": MemoryType.KNOWLEDGE,
|
||||
"技能": MemoryType.SKILL,
|
||||
"目标": MemoryType.GOAL,
|
||||
"经验": MemoryType.EXPERIENCE,
|
||||
"上下文": MemoryType.CONTEXTUAL,
|
||||
}
|
||||
|
||||
|
||||
class ExtractionStrategy(Enum):
|
||||
"""提取策略"""
|
||||
|
||||
@@ -428,7 +446,7 @@ class MemoryBuilder:
|
||||
subject=normalized_subject,
|
||||
predicate=predicate_value,
|
||||
obj=object_value,
|
||||
memory_type=MemoryType(mem_data.get("type", "contextual")),
|
||||
memory_type=self._resolve_memory_type(mem_data.get("type")),
|
||||
chat_id=context.get("chat_id"),
|
||||
source_context=mem_data.get("reasoning", ""),
|
||||
importance=importance_level,
|
||||
@@ -459,7 +477,33 @@ class MemoryBuilder:
|
||||
|
||||
return memories
|
||||
|
||||
def _parse_enum_value(self, enum_cls: type[Enum], raw_value: Any, default: Enum, field_name: str) -> Enum:
|
||||
def _resolve_memory_type(self, type_str: Any) -> MemoryType:
|
||||
"""健壮地解析记忆类型,兼容中文和英文"""
|
||||
if not isinstance(type_str, str) or not type_str.strip():
|
||||
return MemoryType.CONTEXTUAL
|
||||
|
||||
cleaned_type = type_str.strip()
|
||||
|
||||
# 尝试中文映射
|
||||
if cleaned_type in CHINESE_TO_MEMORY_TYPE:
|
||||
return CHINESE_TO_MEMORY_TYPE[cleaned_type]
|
||||
|
||||
# 尝试直接作为枚举值解析
|
||||
try:
|
||||
return MemoryType(cleaned_type.lower().replace(" ", "_"))
|
||||
except ValueError:
|
||||
pass
|
||||
|
||||
# 尝试作为枚举名解析
|
||||
try:
|
||||
return MemoryType[cleaned_type.upper()]
|
||||
except KeyError:
|
||||
pass
|
||||
|
||||
logger.warning(f"无法解析未知的记忆类型 '{type_str}',回退到上下文类型")
|
||||
return MemoryType.CONTEXTUAL
|
||||
|
||||
def _parse_enum_value(self, enum_cls: Type[E], raw_value: Any, default: E, field_name: str) -> E:
|
||||
"""解析枚举值,兼容数字/字符串表示"""
|
||||
if isinstance(raw_value, enum_cls):
|
||||
return raw_value
|
||||
|
||||
Reference in New Issue
Block a user