Update server.py

server.py

@@ -41,19 +41,27 @@ Settings.embed_model = embed_model
 
 
 
-# Configuración inicial
 DOCUMENTS_BASE_PATH = "./"
+RETRIEVERS_JSON_PATH = Path("./retrievers.json")
 
-# Cargar automáticamente las fuentes desde el JSON
+# Cargar metadatos
+def load_retrievers_metadata():
+    try:
+        with open(RETRIEVERS_JSON_PATH, 'r', encoding='utf-8') as f:
+            return json.load(f)
+    except Exception as e:
+        print(f"Error cargando retrievers.json: {str(e)}")
+        return {}
+
+retrievers_metadata = load_retrievers_metadata()
 SOURCES = {source: f"{source.lower()}/" for source in retrievers_metadata.keys()}
 
+# Cargar índices
 indices: Dict[str, VectorStoreIndex] = {}
 
 for source, rel_path in SOURCES.items():
     full_path = os.path.join(DOCUMENTS_BASE_PATH, rel_path)
-    
     if not os.path.exists(full_path):
-        print(f"Advertencia: No se encontró la ruta {full_path} para {source}")
         continue
     
     for root, dirs, files in os.walk(full_path):
@@ -62,10 +70,9 @@ for source, rel_path in SOURCES.items():
                 storage_path = os.path.join(root, "storage_nodes")
                 storage_context = StorageContext.from_defaults(persist_dir=storage_path)
                 index_name = os.path.basename(root)
-                indices[index_name] = index  # Guardamos con nombre directo
+                indices[index_name] = load_index_from_storage(storage_context, index_id="vector_index")
             except Exception as e:
-                print(f"Error cargando índice en {root}: {str(e)}")
-                continue
+                print(f"Error cargando índice {index_name}: {str(e)}")
 
 
 
@@ -109,6 +116,45 @@ def retrieve_docs(
     retrievers: List[str],
     top_k: int = 3
 ) -> dict:
+    """
+    Realiza búsqueda semántica en los documentos indexados.
+    
+    Parámetros:
+        query (str, requerido): Texto de búsqueda. Ejemplo: "vacunas COVID-19"
+        retrievers (List[str], requerido): Lista de nombres de retrievers a consultar. 
+                                          Ejemplo: ["vec_who_1", "fda_tree_1"]
+        top_k (int, opcional): Número máximo de resultados por retriever. Default: 3
+    
+    Retorna:
+        dict: {
+            "query": str,  # Texto buscado
+            "results": {
+                "retriever_name": {
+                    "documents": [
+                        {
+                            "content": str,      # Texto del documento
+                            "metadata": dict,    # Metadatos del documento
+                            "score": float       # Puntaje de relevancia
+                        }
+                    ],
+                    "description": str,  # Descripción del retriever
+                    "source": str        # Fuente (oms, fda, etc.)
+                }
+            },
+            "top_k": int,
+            "warnings": {               # Solo si hay errores
+                "invalid_retrievers": [str],  # Nombres no válidos
+                "valid_options": [str]   # Retrievers disponibles
+            }
+        }
+    
+    Ejemplo de uso:
+        retrieve_docs(
+            query="guías de vacunación",
+            retrievers=["vec_who_2"],
+            top_k=2
+        )
+    """
     results = {}
     invalid = []
     
@@ -121,11 +167,13 @@ def retrieve_docs(
             retriever = indices[name].as_retriever(similarity_top_k=top_k)
             nodes = retriever.retrieve(query)
             
-            # Obtener metadatos del JSON si existen
+            # Obtener metadatos del JSON
             metadata = {}
-            for source in retrievers_metadata.values():
-                if name in source:
-                    metadata = source[name]
+            source = "unknown"
+            for src, indexes in retrievers_metadata.items():
+                if name in indexes:
+                    metadata = indexes[name]
+                    source = src
                     break
             
             results[name] = {
@@ -138,7 +186,7 @@ def retrieve_docs(
                     for node in nodes
                 ],
                 "description": metadata.get("description", ""),
-                "source": next((s for s, idx in retrievers_metadata.items() if name in idx), "unknown")
+                "source": source
             }
         except Exception as e:
             results[name] = {"error": str(e)}