refactor: 去重删除逻辑统一到 VectorDB.delete_by_source

config.yaml

@@ -19,3 +19,5 @@ chunk:
 server:
   host: 0.0.0.0
   port: 8000
+  # ssl_keyfile: ""    # HTTPS 私钥路径（设置后启用 HTTPS）
+  # ssl_certfile: ""   # HTTPS 证书路径（设置后启用 HTTPS）

scripts/serve.py

@@ -1,11 +1,10 @@
 """便捷启动脚本 — 已废弃, 请使用 `uv run md-vector-db serve`."""
+import sys
 import uvicorn
-import warnings
 
-warnings.warn(
-    "scripts/serve.py 已废弃, 请使用 `uv run md-vector-db serve`",
-    DeprecationWarning,
-    stacklevel=2,
+print(
+    "[废弃] scripts/serve.py 已废弃, 请使用 `uv run md-vector-db serve`",
+    file=sys.stderr,
 )
 
 if __name__ == "__main__":

src/core/db.py

@@ -46,6 +46,25 @@ class VectorDB:
             except ValueError:
                 pass  # collection 不存在则忽略
 
+    def delete_by_source(self, collection_name: str, file_name: str) -> bool:
+        """按 source_file 删除文档 (线程安全)."""
+        import logging
+        _logger = logging.getLogger("md-vector-db")
+        collection = self.get_or_create_collection(collection_name)
+        try:
+            with self._write_lock:
+                existing = collection.get(
+                    where={"source_file": file_name}
+                )
+                if existing and existing["ids"]:
+                    collection.delete(ids=existing["ids"])
+                    return True
+        except ValueError:
+            pass
+        except Exception:
+            _logger.exception("删除文档失败: %s (collection=%s)", file_name, collection_name)
+        return False
+
     def close(self) -> None:
         """释放数据库连接."""
         self.client.close()

src/core/ingest.py

@@ -111,15 +111,5 @@ class DocumentIngestor:
         return results
 
     def _remove_by_source(self, file_name: str) -> None:
-        """按 source_file 删除已有 chunks."""
-        try:
-            with self.db.write_lock:
-                existing = self.collection.get(
-                    where={"source_file": file_name}
-                )
-                if existing and existing["ids"]:
-                    self.collection.delete(ids=existing["ids"])
-        except ValueError:
-            pass  # collection 为空时 ChromaDB 抛 ValueError
-        except Exception:
-            logger.exception("去重检查失败: %s", file_name)
+        """按 source_file 删除已有 chunks（委托 VectorDB）."""
+        self.db.delete_by_source(self.collection_name, file_name)

src/core/search.py

@@ -81,17 +81,5 @@ class Searcher:
         return sorted(sources)
 
     def delete_by_source(self, file_name: str) -> bool:
-        """按文件名删除文档 (线程安全)."""
-        try:
-            with self.db.write_lock:
-                existing = self.collection.get(
-                    where={"source_file": file_name}
-                )
-                if existing and existing["ids"]:
-                    self.collection.delete(ids=existing["ids"])
-                    return True
-        except ValueError:
-            pass  # collection 为空时 ChromaDB 抛 ValueError
-        except Exception:
-            logger.exception("删除文档失败: %s", file_name)
-        return False
+        """按文件名删除文档 (委托 VectorDB)."""
+        return self.db.delete_by_source(self.collection_name, file_name)

src/core/splitters/registry.py

@@ -21,7 +21,11 @@ _custom_registry: dict[str, type[Splitter]] = {}
 
 
 def register_splitter(ext: str, splitter_cls: type[Splitter]) -> None:
-    """注册自定义 Splitter 类."""
+    """注册自定义 Splitter 类.
+
+    注意: 此函数非线程安全，请在程序启动时调用（单线程阶段）。
+    运行时动态注册需自行加锁。
+    """
     ext = ext.lower() if ext.startswith(".") else f".{ext}"
     _custom_registry[ext] = splitter_cls