接口报错异常处理

frontend-web/src/open-canvas/components/ai-writing/TimelineStep.tsx

@@ -77,6 +77,17 @@ export function TimelineStep({ event, isLast, children }: TimelineStepProps) {
                 {event.detail}
               </p>
             )}
+            {/* 接口解析失败时的原始错误响应体，默认折叠，点击展开 */}
+            {event.raw && (
+              <details className="mt-1">
+                <summary className="cursor-pointer select-none text-[11px] font-medium text-muted-foreground hover:text-foreground">
+                  错误返回结果（点击展开）
+                </summary>
+                <pre className="mt-1 max-h-60 overflow-auto whitespace-pre-wrap break-all rounded bg-muted/60 p-2 text-[10px] leading-relaxed text-muted-foreground">
+                  {event.raw}
+                </pre>
+              </details>
+            )}
           </div>
         ) : (
           <span className="ml-1.5 text-xs text-muted-foreground">{event.message}</span>

frontend-web/src/open-canvas/types/ai-writing.ts

@@ -90,6 +90,8 @@ export interface ProgressEvent {
   // notice（非致命告警）/ error：标题 + 真实原因
   title?: string;
   detail?: string;
+  // 接口解析失败时的原始错误响应体，前端默认折叠展示
+  raw?: string;
   // materials_ready
   materials?: AIMaterial[];
   keywords?: string[];

offline-backend-20260512/backend/packages/harness/deerflow/agents/ai_writing/nodes/researcher.py

@@ -21,9 +21,15 @@ logger = logging.getLogger(__name__)
 _MAX_MATERIALS = 12
 
 
-def _notice(title: str, detail: str) -> dict:
+def _notice(title: str, detail: str, raw: str = "") -> dict:
-    """构造一个非致命告警事件，前端在时间线显示「标题 + 真实原因」，不打断写作流程。"""
+    """构造一个非致命告警事件，前端在时间线显示「标题 + 真实原因」，不打断写作流程。
-    return {"type": "notice", "agent_name": "素材收集专家", "title": title, "detail": detail}
+
+    raw 非空时附带接口的原始错误响应体，前端默认折叠展示。
+    """
+    event = {"type": "notice", "agent_name": "素材收集专家", "title": title, "detail": detail}
+    if raw:
+        event["raw"] = raw
+    return event
 
 
 async def researcher_node(state: AIWritingState, config: RunnableConfig) -> dict:
@@ -173,7 +179,13 @@ async def researcher_node(state: AIWritingState, config: RunnableConfig) -> dict
         notices.append(_notice("模型错误", f"素材排序失败，已按相关度兜底排序。原因：{rank_error}"))
     if search_errors:
         uniq = list(dict.fromkeys(search_errors))
-        notices.append(_notice("ES 波动导致检索异常", "；".join(uniq[:3])))
+        # 解析失败时收集到的接口原始响应体，前端折叠展示便于定位
+        raw_dumps = list(dict.fromkeys(getattr(provider, "raw_responses", []) or []))
+        notices.append(_notice(
+            "ES 波动导致检索异常",
+            "；".join(uniq[:3]),
+            raw="\n\n----------\n\n".join(raw_dumps[:3]),
+        ))
 
     return {
         "material_package": pkg,

offline-backend-20260512/backend/packages/harness/deerflow/agents/ai_writing/search/intranet_stub.py

@@ -11,6 +11,8 @@ logger = logging.getLogger(__name__)
 # 内网知识库默认检索接口与知识库名（可经 config.yaml 的 ai_writing 段覆盖）
 _DEFAULT_ENDPOINT = "http://10.106.57.180:7863/knowledge_base/search_docs_es"
 _DEFAULT_KB = "samples"
+# 单条原始响应体最大保留长度，避免把超大响应整段灌进前端时间线
+_RAW_SNIPPET_LIMIT = 2000
 
 
 class IntranetSearchProvider(BaseSearchProvider):
@@ -25,14 +27,26 @@ class IntranetSearchProvider(BaseSearchProvider):
         self,
         base_url: str = "",
         knowledge_base_name: str = _DEFAULT_KB,
-        timeout: int = 15,
+        timeout: int = 120,
         **_ignored,
     ):
         self.endpoint = base_url or _DEFAULT_ENDPOINT
         self.knowledge_base_name = knowledge_base_name or _DEFAULT_KB
+        # 内网 ES 接口可能较慢，超时放宽到 2 分钟
         self.timeout = timeout
         # 收集本次运行的检索异常，供 researcher 节点取出并提示用户（ES 波动等）
         self.errors: list[str] = []
+        # 解析失败时收集的原始响应体，供前端「错误返回结果」折叠展示
+        self.raw_responses: list[str] = []
+
+    def _record_raw(self, body: str) -> None:
+        """收集解析失败 / 结构异常时的原始响应体，供前端折叠查看。"""
+        body = (body or "").strip()
+        if not body:
+            return
+        if len(body) > _RAW_SNIPPET_LIMIT:
+            body = body[:_RAW_SNIPPET_LIMIT] + "…（响应体过长，已截断）"
+        self.raw_responses.append(body)
 
     async def search(self, query: str, max_results: int = 5, **kwargs) -> List[SearchResult]:
         import httpx
@@ -42,21 +56,38 @@ class IntranetSearchProvider(BaseSearchProvider):
             "knowledge_base_name": self.knowledge_base_name,
             "ts": "yes",
         }
+        # 先拿到原始响应文本，JSON 解析失败时把它原样回传前端便于定位接口返回了什么
+        raw_text = ""
         try:
             async with httpx.AsyncClient(timeout=self.timeout) as client:
                 resp = await client.post(self.endpoint, json=payload)
                 resp.raise_for_status()
+                raw_text = resp.text
                 data = resp.json()
-        except Exception as e:
+        except httpx.HTTPStatusError as e:
+            # 非 2xx：接口有响应但状态码异常，把响应体一并暴露给前端
             logger.warning("intranet search failed for query=%r: %s", query, e)
             self.errors.append(f"{type(e).__name__}: {e}")
+            self._record_raw(e.response.text if e.response is not None else "")
+            return []
+        except Exception as e:
+            # 连接失败 / 超时 / JSON 解析失败等；JSON 解析失败时 raw_text 已是接口响应体
+            logger.warning("intranet search failed for query=%r: %s", query, e)
+            self.errors.append(f"{type(e).__name__}: {e}")
+            self._record_raw(raw_text)
             return []
 
-        # 接口直接返回数组；兼容个别情况下包了一层 data/results 的情况
+        # 正常情况下接口直接返回数组；兼容个别情况下包了一层 data/results 的结构
-        if isinstance(data, dict):
+        if isinstance(data, list):
+            items = data
+        elif isinstance(data, dict):
             items = data.get("data") or data.get("results") or []
         else:
-            items = data or []
+            # 返回结构既不是数组也不是已知 dict 包装：视为异常，把原始响应回传前端
+            logger.warning("intranet search unexpected response type for query=%r: %s", query, type(data).__name__)
+            self.errors.append(f"返回结构异常：期望数组，实际为 {type(data).__name__}")
+            self._record_raw(raw_text)
+            return []
 
         results: List[SearchResult] = []
         for i, item in enumerate(items[:max_results]):