Se hace una depuración inicial del proyecto

agents/app/langgraph_tools/tools/orders/order_tools_2.py

@@ -133,7 +133,6 @@ def search_sale_order(order_id: int):
 
     return response_sale
 
-
 @tool
 def add_lines_to_order(
     order_id: int, product: str, unit: str, quantity: str, unitprice: str
@@ -210,36 +209,3 @@ tools = [
     add_lines_to_order,
     search_associate_party_to_contact_mechanism,
 ]
-
-# if __name__ == "__main__":
-#     # Crear una orden de venta
-#     party = 2573
-#     pickup_location = "at_home"
-#     order_id = create_sale_order(party=party, pickup_location=pickup_location)
-#     print(f"\nOrden creada con ID: {order_id}")
-
-#     # Agregar líneas a la orden
-#     product = "1"
-#     unit = "1"
-#     quantity = "3"
-#     unitprice = "15"
-
-#     add_line_response = add_lines_to_order(order_id, product, unit, quantity, unitprice)
-#     print(f"\nRespuesta al agregar línea: {add_line_response.text}")
-
-#     # Verificar la orden actualizada
-#     updated_order = search_sale_order(order_id)
-#     print(f"\nOrden actualizada: {updated_order.text}")
-
-#     # Agregar otra línea a la orden
-#     product = "2"
-#     unit = "1"
-#     quantity = "3"
-#     unitprice = "15"
-
-#     add_line_response = add_lines_to_order(order_id, product, unit, quantity, unitprice)
-#     print(f"\nRespuesta al agregar línea: {add_line_response.text}")
-
-#     # Verificar la orden actualizada
-#     updated_order = search_sale_order(order_id)
-#     print(f"\nOrden actualizada: {updated_order.text}")

agents/app/prueba_catalog.py

@@ -1,82 +0,0 @@
-from seller.catalog_tools import CatalogTrytonTools
-import json
-
-# result = json.dumps(response, indent=4)
-
-url = "http://192.168.0.25:8000"
-key = "9a9ffc430146447d81e6698240199a4be2b0e774cb18474999d0f60e33b5b1eb1cfff9d9141346a98844879b5a9e787489c891ddc8fb45cc903b7244cab64fb1"
-db = "tryton"
-application_name = "sale_don_confiao"
-
-
-catalog = CatalogTrytonTools(url, application_name, key, db)
-
-# Lista de los productos
-print("=== Productos ===")
-
-response = catalog.list_products()
-
-products: str = ""
-
-for product in response:
-    id = product["id"]
-    name = product["name"]
-    id_unit_measurement = product["default_uom."]["id"]
-    products += f"- id: {id} - Nombre: {name} - ID Unidad de Medida: {id_unit_measurement}\n"
-
-print(products)
-
-# Precio del los productos 
-print("=== Precios de los productos ===")
-
-response = catalog.search_products("Arroz")
-precios: str = ""
-
-for product in response:
-    id = product["id"]
-    name = product["name"]
-    precio = product["template."]["list_price"]["decimal"]
-    id_unit_measurement = product["default_uom."]["id"]
-
-    precios += f"- id: {id} - Nombre: {name}: ${precio} - ID Unidad de Medida: {id_unit_measurement}\n"
-
-print(precios)
-
-# Revisar disponibilidad de los productos
-# print("=== Disponibilidad de los productos ===")
-
-# response = catalog.search_products("Papa")
-# disponibilidad: str = ""
-
-# for product in response:
-#     id = product["id"]
-#     name = product["name"]
-#     stock = product["quantity"]
-
-#     disponibilidad += f"- id: {id} - Nombre: {name} - Stock: {stock}\n"
-
-# print(disponibilidad)
-
-
-
-# from langgraph_tools.tools.catalog.catalog_tools import (
-#     list_products,
-#     search_products,
-#     check_price,
-# )
-
-# def test_catalog_functions():
-#     print("=== Probando lista de productos ===")
-#     print(list_products())
-#     print("\n")
-
-#     print("=== Probando búsqueda de productos ===")
-#     print(search_products("Arroz"))
-#     print("\n")
-
-#     print("=== Probando verificación de precio ===")
-#     print(check_price("Arroz"))
-#     print("\n")
-
-# if __name__ == "__main__":
-#     test_catalog_functions()

agents/app/prueba_llm.py

@@ -1,20 +0,0 @@
-# from langchain_tools.llm import load_llm_openai
-# import yaml
-#
-# with open("langgraph_tools/prompts.yaml", "r") as f:
-#     PROMPTS = yaml.safe_load(f)
-#
-#
-# llm = load_llm_openai()
-# query = "Necesito informacion sobre la tienda."
-# prompt = PROMPTS["classifier"]["system"].format(query=query)
-#
-# response: dict = llm.invoke(prompt)
-#
-# print(response.content)
-
-from langgraph_tools.tools.general_info import get_link_page
-
-
-link = get_link_page()
-print(link)

agents/app/rag/embeddings.py

@@ -1,10 +0,0 @@
-from dotenv import load_dotenv
-from langchain_openai import OpenAIEmbeddings
-
-
-def load_embeddins():
-    load_dotenv()
-    # model = "text-embedding-ada-002"
-    model = "text-embedding-3-small"
-
-    return OpenAIEmbeddings(model=model)

agents/app/rag/llm.py

@@ -1,17 +0,0 @@
-from dotenv import load_dotenv
-from langchain_openai import ChatOpenAI
-
-
-def load_llm_openai():
-    load_dotenv()
-    # model = "gpt-3.5-turbo-0125"
-    # model = "gpt-4o"
-    model = "gpt-4o-mini"
-
-    llm = ChatOpenAI(
-        model=model,
-        temperature=0.1,
-        max_tokens=2000,
-    )
-
-    return llm

agents/app/rag/rag_chain.py

@@ -1,46 +0,0 @@
-from langchain.chains import create_history_aware_retriever
-from langchain_core.prompts import ChatPromptTemplate, MessagesPlaceholder
-from langchain.chains import create_retrieval_chain
-from langchain.chains.combine_documents import create_stuff_documents_chain
-
-
-def create_rag_chain(llm, retriever):
-    contextualize_q_system_prompt = """
-    Given a chat history and the latest user question \
-    which might reference context in the chat history,
-    formulate a standalone question \
-    which can be understood without the chat history.
-    Do NOT answer the question, \
-    just reformulate it if needed and otherwise return it as is.
-    """
-    contextualize_q_prompt = ChatPromptTemplate.from_messages(
-        [
-            ("system", contextualize_q_system_prompt),
-            MessagesPlaceholder("chat_history"),
-            ("human", "{input}"),
-        ]
-    )
-    history_aware_retriever = create_history_aware_retriever(
-        llm, retriever, contextualize_q_prompt
-    )
-
-    # ___________________Chain con el chat history_______________________-
-    qa_system_prompt = """
-    You are an assistant for question-answering tasks.  \
-    Use the following pieces of retrieved context to answer the question. \
-    If you don't know the answer, just say that you don't know. \
-    The length of the answer should be sufficient to address
-    what is being asked, \
-    meaning don't limit yourself in length.\
-    {context}"""
-    qa_prompt = ChatPromptTemplate.from_messages(
-        [
-            ("system", qa_system_prompt),
-            MessagesPlaceholder("chat_history"),
-            ("human", "{input}"),
-        ]
-    )
-    question_answer_chain = create_stuff_documents_chain(llm, qa_prompt)
-
-    return create_retrieval_chain(
-        history_aware_retriever, question_answer_chain)

agents/app/rag/retriever.py

@@ -1,18 +0,0 @@
-from langchain_chroma import Chroma
-
-
-def create_retriever(embeddings, persist_directory: str):
-    # Cargamos la vectorstore
-    # vectordb = Chroma.from_documents(
-    #     persist_directory=st.session_state.persist_directory,
-    # Este es el directorio del la vs del docuemnto del usuario
-    # que se encuentra cargado en la session_state.
-    #     embedding_function=embeddings,
-    # )
-    vectordb = Chroma(
-        persist_directory=persist_directory,
-        embedding_function=embeddings,
-    )
-
-    # Creamos el retriver para que retorne los fragmentos mas relevantes.
-    return vectordb.as_retriever(search_kwargs={"k": 6})

agents/app/rag/split_docs.py

@@ -1,42 +0,0 @@
-from langchain_community.document_loaders.pdf import PyPDFLoader
-from langchain.text_splitter import RecursiveCharacterTextSplitter
-import os
-
-
-# def load_split_docs(file_name: str) -> list:
-#     file_path: str = os.path.join("documents", "pdfs", file_name)
-#     loader = PyPDFLoader(file_path)
-#     docs: list = loader.load()
-#     chunk_size: int = 2000
-#     chunk_overlap: int = 300
-#
-#     splitter = RecursiveCharacterTextSplitter(
-#         chunk_size=chunk_size, chunk_overlap=chunk_overlap
-#     )
-#     docs_split: list = splitter.split_documents(docs)
-#
-#     return docs_split
-
-
-def load_split_docs(file_name: str) -> list:
-    # Obtener el directorio base del proyecto
-    base_dir = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
-
-    # Construir la ruta absoluta al PDF
-    file_path = os.path.join(base_dir, "documents", "pdfs", file_name)
-
-    # Verificar si el archivo existe
-    if not os.path.exists(file_path):
-        print(f"Archivo no encontrado en: {file_path}")
-        raise FileNotFoundError(f"No se encontró el archivo en: {file_path}")
-
-    loader = PyPDFLoader(file_path)
-    docs: list = loader.load()
-
-    chunk_size: int = 2000
-    chunk_overlap: int = 300
-    splitter = RecursiveCharacterTextSplitter(
-        chunk_size=chunk_size, chunk_overlap=chunk_overlap
-    )
-    docs_split: list = splitter.split_documents(docs)
-    return docs_split

agents/app/rag/vectorstore.py

@@ -1,48 +0,0 @@
-from langchain_chroma import Chroma
-import os
-
-#
-# def create_vectorstore(docs_split: list, embeddings, file_name: str):
-#     db_name: str = file_name.replace(".pdf", "").replace(" ", "_").lower()
-#     persist_directory: str = f"embeddings/{db_name}"
-#
-#     # Crear el directorio si no existe
-#     os.makedirs(persist_directory, exist_ok=True)
-#
-#     # Siempre crear/actualizar el vectorstore
-#     vectordb = Chroma.from_documents(
-#         persist_directory=persist_directory,
-#         documents=docs_split,
-#         embedding=embeddings,
-#     )
-#
-#     return vectordb
-
-
-def create_vectorstore(docs_split: list, embeddings, file_name: str):
-    # Obtener el directorio base del proyecto
-    base_dir = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
-
-    # Crear el nombre de la base de datos
-    db_name: str = file_name.replace(".pdf", "").replace(" ", "_").lower()
-
-    # Construir la ruta absoluta para los embeddings
-    persist_directory: str = os.path.join(base_dir, "embeddings", db_name)
-
-    # Crear el directorio si no existe
-    os.makedirs(persist_directory, exist_ok=True)
-
-    # Debug log
-    print(f"Creando vectorstore en: {persist_directory}")
-
-    try:
-        # Crear/actualizar el vectorstore
-        vectordb = Chroma.from_documents(
-            persist_directory=persist_directory,
-            documents=docs_split,
-            embedding=embeddings,
-        )
-        return vectordb
-    except Exception as e:
-        print(f"Error al crear vectorstore: {e}")
-        raise

agents/test_create_party.py

@@ -1,27 +0,0 @@
-#!/usr/bin/env python3
-from app.langgraph_tools.tools.orders.order_tools_2 import create_party
-import json
-
-# Prueba de la función create_party
-def test_create_party():
-    # Parámetros de ejemplo
-    party_full_name = "Cristian Garces"
-    contact_method_type = "email"
-    contact_method_value = "cristian.garces@example.com"
-    
-    # Llamar a la función
-    response = create_party(
-        party_full_name=party_full_name,
-        contact_method_type=contact_method_type,
-        contact_method_value=contact_method_value
-    )
-    
-    # Imprimir la respuesta
-    print("Status Code:", response.status_code)
-    try:
-        print("Response:", json.loads(response.text))
-    except json.JSONDecodeError:
-        print("Raw Response:", response.text)
-
-if __name__ == "__main__":
-    test_create_party()