fix simulation reporting and research resilience

2026-09-28 15:43:23 +08:00
parent 23297be2d7
commit a428214c53
25 changed files with 1488 additions and 244 deletions
@@ -387,12 +387,18 @@ class CitizenGenerator:
        )

        data = parse_json(response, purpose="居民名单生成")
-        if isinstance(data, dict) and "citizens" in data:
-            cast = data["citizens"]
-        elif isinstance(data, list):
-            cast = data
-        else:
-            cast = []
+        if isinstance(data, dict) and "citizens" in data:
+            cast = data["citizens"]
+        elif isinstance(data, list):
+            cast = data
+        else:
+            cast = []
+        cast = [
+            member for member in cast
+            if isinstance(member, dict)
+            and str(member.get("name", "")).strip()
+            and str(member.get("role", "")).strip()
+        ]

        # If the user picked a Chinese locale but the LLM ignored our language
        # directive and returned English names, retry once with a stronger nudge.
@@ -414,33 +420,74 @@ class CitizenGenerator:
                purpose="cast_generate_retry",
            )
            data = parse_json(response, purpose="居民名单生成重试")
-            if isinstance(data, dict) and "citizens" in data:
-                cast = data["citizens"]
-            elif isinstance(data, list):
-                cast = data
-
-        if len(cast) < count:
-            logger.warning("居民生成仅返回 %d/%d 人，将使用默认居民补齐", len(cast), count)
-            while len(cast) < count:
-                idx = len(cast)
-                if self._locale == "zh":
-                    cast.append({
-                        "name": f"居民{idx + 1}号",
-                        "role": "劳动者",
-                        "age": 30 + (idx % 40),
-                        "personality_hook": "一个安静过日子的人",
-                        "stance": "对规则漠不关心",
-                    })
-                else:
-                    cast.append({
-                        "name": f"Citizen {idx + 1}",
-                        "role": "Laborer",
-                        "age": 30 + (idx % 40),
-                        "personality_hook": "A quiet person trying to get by",
-                        "stance": "Indifferent to the rules",
-                    })
-
-        return cast[:count]
+            if isinstance(data, dict) and "citizens" in data:
+                cast = data["citizens"]
+            elif isinstance(data, list):
+                cast = data
+            cast = [
+                member for member in cast
+                if isinstance(member, dict)
+                and str(member.get("name", "")).strip()
+                and str(member.get("role", "")).strip()
+            ]
+
+        if len(cast) < count:
+            missing = count - len(cast)
+            refill_user = user_text + (
+                f"\n\n[REFILL] You returned {len(cast)} citizens. Generate exactly {missing} additional "
+                "citizens only. They must be distinct, neutral, and contain all required fields."
+            )
+            if self._locale == "zh":
+                refill_user += " Use Chinese names, roles, hooks, and stances."
+            try:
+                refill_response = await self.llm.generate(
+                    system=system_prompt,
+                    user=refill_user,
+                    json_mode=True,
+                    max_tokens=min(3000, max(1000, missing * 220)),
+                    purpose="cast_generate_refill",
+                )
+                refill_data = parse_json(refill_response, purpose="居民名单补生成")
+                if isinstance(refill_data, dict) and isinstance(refill_data.get("citizens"), list):
+                    refill_cast = refill_data["citizens"]
+                elif isinstance(refill_data, list):
+                    refill_cast = refill_data
+                else:
+                    refill_cast = []
+                cast.extend(
+                    member for member in refill_cast
+                    if isinstance(member, dict)
+                    and str(member.get("name", "")).strip()
+                    and str(member.get("role", "")).strip()
+                )
+                cast = cast[:count]
+            except Exception as e:
+                logger.warning("居民名单补生成失败：%s", e)
+
+        if len(cast) < count:
+            logger.error("居民生成仍不足 %d/%d 人，将进入带标记的中性降级模式", len(cast), count)
+            while len(cast) < count:
+                idx = len(cast)
+                if self._locale == "zh":
+                    cast.append({
+                        "name": f"待生成居民{idx + 1}",
+                        "role": "未指定",
+                        "age": 30 + (idx % 40),
+                        "personality_hook": "个体化背景尚未生成",
+                        "stance": "暂无明确立场",
+                        "_generation_source": "fallback",
+                    })
+                else:
+                    cast.append({
+                        "name": f"Pending Citizen {idx + 1}",
+                        "role": "Unspecified",
+                        "age": 30 + (idx % 40),
+                        "personality_hook": "Individual background was not generated",
+                        "stance": "No stated position yet",
+                        "_generation_source": "fallback",
+                    })
+
+        return cast[:count]

    async def _generate_personas(
        self,
@@ -461,10 +508,10 @@ class CitizenGenerator:
                    f"Resources: {', '.join(blueprint.resources)}\n"
                    f"Tensions: {'; '.join(blueprint.initial_tensions)}"
                )
-                citizen_ctx = (
-                    f"Name: {member['name']}\n"
-                    f"Role: {member['role']}\n"
-                    f"Age: {member['age']}\n"
+                citizen_ctx = (
+                    f"Name: {member.get('name', 'Unnamed Citizen')}\n"
+                    f"Role: {member.get('role', 'Unspecified')}\n"
+                    f"Age: {member.get('age', 30)}\n"
                    f"Personality hook: {member.get('personality_hook', '')}\n"
                    f"Stance on tensions: {member.get('stance', 'unknown')}"
                )
@@ -495,9 +542,13 @@ class CitizenGenerator:
                idx = batch_start + i
                member = cast[idx]
                data = parse_json(response, purpose=f"居民画像生成（批次索引={batch_start + i}）")
-
-                personality_data = data.get("personality", {})
-                personality = Personality(
+                if not isinstance(data, dict):
+                    data = {}
+
+                personality_data = data.get("personality", {})
+                if not isinstance(personality_data, dict):
+                    personality_data = {}
+                personality = Personality(
                    honesty=self._clamp(personality_data.get("honesty", 0.5)),
                    ambition=self._clamp(personality_data.get("ambition", 0.5)),
                    empathy=self._clamp(personality_data.get("empathy", 0.5)),
@@ -509,19 +560,26 @@ class CitizenGenerator:
                    novelty_seeking=self._clamp(personality_data.get("novelty_seeking", 0.5)),
                )

-                resources = data.get("resources", {})
-                for res in blueprint.resources:
+                resources = data.get("resources", {})
+                if not isinstance(resources, dict):
+                    resources = {}
+                for res in blueprint.resources:
                    if res not in resources:
                        resources[res] = 50

-                raw_style = member.get("communication_style", "").lower().strip()
+                raw_style = str(member.get("communication_style", "") or "").lower().strip()
                comm_style = raw_style if raw_style in COMMUNICATION_STYLES else random.choice(list(COMMUNICATION_STYLES.keys()))

-                agent = AgentPersona(
-                    id=idx,
-                    name=member["name"],
-                    role=member["role"],
-                    age=member.get("age", 30),
+                try:
+                    age = max(18, min(75, int(member.get("age", 30))))
+                except (TypeError, ValueError):
+                    age = 30
+
+                agent = AgentPersona(
+                    id=idx,
+                    name=str(member.get("name", "Unnamed Citizen")),
+                    role=str(member.get("role", "Unspecified")),
+                    age=age,
                    personality=personality,
                    background=data.get("background", member.get("personality_hook", "")),
                    goals=data.get("goals", ["Survive", "Find purpose"]),
@@ -533,8 +591,9 @@ class CitizenGenerator:
                    location="community",
                    faction=None,
                    emotional_state="curious",
-                    communication_style=comm_style,
-                )
+                    communication_style=comm_style,
+                    generation_source=str(member.get("_generation_source", "llm")),
+                )

                agents.append(agent)
                if on_citizen: