refactor: brain memory

metagpt/actions/talk_action.py

@@ -6,10 +6,12 @@
 @File    : talk_action.py
 @Desc    : Act as it’s a talk
 """
+import json
 
 from metagpt.actions import Action, ActionOutput
 from metagpt.config import CONFIG
 from metagpt.const import DEFAULT_LANGUAGE
+from metagpt.llm import LLMType
 from metagpt.logs import logger
 
 
@@ -63,6 +65,15 @@ class TalkAction(Action):
         return prompt
 
     async def run(self, *args, **kwargs) -> ActionOutput:
+        if CONFIG.LLM_TYPE == LLMType.METAGPT.value:
+            rsp = await self.llm.aask(
+                msg=self._talk,
+                knowledge_msgs=[{"knowledge": self._knowledge}] if self._knowledge else None,
+                history_msgs=json.loads(self._history_summary) if self._history_summary else None,
+            )
+            self._rsp = ActionOutput(content=rsp)
+            return self._rsp
+
         prompt = self.prompt
         rsp = await self.llm.aask(msg=prompt, system_msgs=[])
         logger.debug(f"PROMPT:{prompt}\nRESULT:{rsp}\n")

metagpt/memory/brain_memory.py

@@ -309,4 +309,28 @@ class BrainMemory(pydantic.BaseModel):
     def is_history_available(self):
         return bool(self.history or self.historical_summary)
 
+    @property
+    def history_text(self):
+        if self.llm_type == LLMType.METAGPT.value:
+            return self._get_metagpt_history_text()
+        return self._get_openai_history_text()
+
+    def _get_metagpt_history_text(self):
+        return BrainMemory.to_metagpt_history_format(self.history)
+
+    def _get_openai_history_text(self):
+        if len(self.history) == 0 and not self.historical_summary:
+            return ""
+        texts = [self.historical_summary] if self.historical_summary else []
+        for m in self.history[:-1]:
+            if isinstance(m, Dict):
+                t = Message(**m).content
+            elif isinstance(m, Message):
+                t = m.content
+            else:
+                continue
+            texts.append(t)
+
+        return "\n".join(texts)
+
     DEFAULT_TOKEN_SIZE = 500

metagpt/provider/base_gpt_api.py

@@ -38,11 +38,22 @@ class BaseGPTAPI(BaseChatbot):
         rsp = self.completion(message)
         return self.get_choice_text(rsp)
 
-    async def aask(self, msg: str, system_msgs: Optional[list[str]] = None, generator: bool = False) -> str:
+    async def aask(
+        self,
+        msg: str,
+        system_msgs: Optional[list[str]] = None,
+        history_msgs: Optional[list[dict[str, str]]] = None,
+        knowledge_msgs: Optional[list[dict[str, str]]] = None,
+        generator: bool = False,
+    ) -> str:
+        message = []
         if system_msgs:
-            message = self._system_msgs(system_msgs) + [self._user_msg(msg)]
-        else:
-            message = [self._default_system_msg(), self._user_msg(msg)]
+            message = self._system_msgs(system_msgs)
+        if knowledge_msgs:
+            message.extend(knowledge_msgs)
+        if history_msgs:
+            message.extend(history_msgs)
+        message.append(self._user_msg(msg))
         try:
             rsp = await self.acompletion_text(message, stream=True, generator=generator)
         except Exception as e: