Update core/pipeline_agent.py

core/pipeline_agent.py

@@ -136,12 +136,20 @@ class SSciPipelineAgent:
     @retry(stop=stop_after_attempt(3), wait=wait_random_exponential(multiplier=1, max=10), retry=retry_if_exception_type((requests.exceptions.Timeout, requests.exceptions.ConnectionError, requests.exceptions.HTTPError)))
     def _requests_get_with_retry(self, url, headers=None, params=None, timeout=15):
         resp = requests.get(url, headers=headers, params=params, timeout=timeout)
+        
         if resp.status_code == 429:
             logging.warning(f"⚠️ [API 限流] 触发 429，当前线程进入随机休眠退避。URL: {url}")
             resp.raise_for_status()
+            
+        # 🚨 核心修复：豁免 S2 因为翻页越界导致的 400 错误，保护引擎不被熔断
+        if resp.status_code == 400 and "not available" in resp.text:
+            logging.info(f"ℹ️ [API 翻页边界] 该关键词搜索结果已触底，停止翻页。")
+            return resp
+            
         if resp.status_code != 200:
             logging.error(f"❌ [API 错误] 状态码: {resp.status_code}, 响应: {resp.text}")
             resp.raise_for_status()
+            
         return resp
 
     def _is_english_dominant(self, text):
@@ -380,7 +388,6 @@ class SSciPipelineAgent:
                 
                 if not papers: raise ValueError("Extracted JSON is null or invalid")
                 
-                # 修复可能包裹在 dict 中的情况 (e.g., {"papers": [...]})
                 if isinstance(papers, dict):
                     for v in papers.values():
                         if isinstance(v, list):
@@ -563,7 +570,7 @@ class SSciPipelineAgent:
                                 else: author_str = formatted_authors[0]
 
                                 apa_full = f"{author_str} ({year_str}). {title}. {venue}."
-                                if doi: apa_full += f" [https://doi.org/](https://doi.org/){doi}"
+                                if doi: apa_full += f" https://doi.org/{doi}"
 
                                 with self.lock:
                                     cite_tag = f"[CITE-{len(self.citation_map)+1:02d}]"
@@ -622,7 +629,7 @@ class SSciPipelineAgent:
                                 seen_identifiers.add(identifier)
 
                                 venue = ((item.get("primary_location") or {}).get("source") or {}).get("display_name", "Journal")
-                                doi_clean = (item.get("doi") or "").replace("[https://doi.org/](https://doi.org/)", "")
+                                doi_clean = (item.get("doi") or "").replace("https://doi.org/", "")
                                 formatted_authors, last_names = [], []
                                 for a in item.get("authorships", []):
                                     name_raw = a.get("author", {}).get("display_name", "")
@@ -646,7 +653,7 @@ class SSciPipelineAgent:
                                 else: author_str = formatted_authors[0]
                                 
                                 apa_full = f"{author_str} ({year_str}). {title}. {venue}."
-                                if doi_clean: apa_full += f" [https://doi.org/](https://doi.org/){doi_clean}"
+                                if doi_clean: apa_full += f" https://doi.org/{doi_clean}"
 
                                 with self.lock:
                                     cite_tag = f"[CITE-{len(self.citation_map)+1:02d}]"