refactor: refactor repo structure for better expandability (#27)

* refactor: Update import paths for LlamaCppClient and related modules

* refactor: Restructure repository for improved modularity and clarity

* refactor: Restructure repository for improved modularity and clarity

* refactor: Update test coverage paths in CI configuration

* refactor: Rename functions and variables for clarity and consistency

Author: umbertogriffo

.github/workflows/ci.yaml

@@ -84,4 +84,4 @@ jobs:
 
       - name: Run tests
         run: |
-          poetry run python -m pytest --durations=5 --log-cli-level=DEBUG --capture=tee-sys -v --cov=chatbot --cov-report=term-missing --cov-fail-under=60
+          poetry run python -m pytest --durations=5 --log-cli-level=DEBUG --capture=tee-sys -v --cov=backend --cov-report=term-missing --cov-fail-under=60

Makefile

@@ -15,7 +15,7 @@ install_pre_commit:
 	poetry run pre-commit install --hook-type pre-commit
 
 migrate_db:
-	cd backend && PYTHONPATH=.:../chatbot poetry run python migration.py
+	cd backend && poetry run python migration.py
 
 start_llama_server_cuda:
 	docker compose up -d

README.md

@@ -114,11 +114,13 @@ To easily install the dependencies and start the services we created a make file
     * Both starts `llama.cpp` server locally via Docker compose.
 * Start: ```make start```
     *  Start both the backend and frontend ensuring that the backend is running and ready before launching the frontend.
-* Stop `llama.cpp` Server: ```make stop_llama_server```
-    * Stop the llama.cpp server if it's running locally.
-* Start Server: ```make start_server```
+* Start llama.cpp server
+    * on CUDA: ```make start_llama_server_cuda```
+    * on Metal: ```make start_llama_server_metal```
     * Start the llama.cpp server locally via Docker compose.
     * It will be available at http://0.0.0.0:8080 (it will show the llama-ui).
+* Stop `llama.cpp` Server: ```make stop_llama_server```
+    * Stop the llama.cpp server if it's running locally.
 * Update: ```make update```
     * Update an environment and installs all updated dependencies.
 * Tidy up the code: ```make tidy```
@@ -240,25 +242,14 @@ Build the memory index by running:
 
 ```shell
 make migrate_db
-python chatbot/memory_builder.py --model-name jinaai/jina-embeddings-v5-text-small-retrieval --chunk-size 1000 --chunk-overlap 50
+cd scripts && PYTHONPATH=.:../backend python memory_builder.py --model-name jinaai/jina-embeddings-v5-text-small-retrieval --chunk-size 1000 --chunk-overlap 50
 ```
 
 ## Run the Chatbot
 
 The Chatbot has a UI built with `Vite`, `React` and `TypeScript`, and a backend built with `FastAPI` that serves the LLMs through `llama.cpp` server.
 
-To start the backend type:
-
-```shell
-cd backend && PYTHONPATH=.:../chatbot uvicorn main:app --reload
-```
-
-To start the frontend (in a new terminal):
-```shell
-cd frontend && yarn dev
-```
-
-or to start both ensuring that the backend is running and ready before launching the frontend just run:
+To start both the backend and frontend ensuring that the backend is running and ready before launching the frontend just run:
 
 ```shell
 make start

backend/README.md

@@ -0,0 +1,5 @@
+To start the backend type:
+
+```shell
+cd backend && uvicorn main:app --reload
+```

backend/alembic/env.py

@@ -1,8 +1,7 @@
 from logging.config import fileConfig
 
 from alembic import context
-from bot.memory.document_registry import DocumentRecord  # noqa: F401 – registers the model
-from core.config import settings
+from config import settings
 from helpers.log import get_logger
 from sqlalchemy import engine_from_config, pool
 from sqlmodel import SQLModel

backend/api/deps.py

@@ -5,11 +5,10 @@
 from typing import Annotated, Generator
 
 import state
-from bot.client.llamacpp_client import LlamaCppClient
-from bot.conversation.chat_history import ChatHistory
-from bot.memory.vector_database.chroma import Chroma
-from chat_history import chat_history
+from chat_history import ChatHistory, chat_history
 from fastapi import Depends
+from llm_providers.llamacpp_client import LlamaCppClient
+from memory.vector_database.chroma import Chroma
 from sqlmodel import Session
 
 
@@ -31,14 +30,14 @@ def get_index() -> Generator[Chroma, None, None]:
     """
     Dependency to get the vector database index instance.
     """
-    yield state.index
+    yield state.vector_database
 
 
 def get_db_session() -> Generator[Session, None, None]:
     """
     Create a new database session and close the session after the operation has ended.
     """
-    with Session(state.engine) as session:
+    with Session(state.db_engine) as session:
         yield session
 
 

backend/api/endpoints/chat.py

@@ -1,4 +1,4 @@
-from core.config import settings
+from config import settings
 from fastapi import APIRouter
 from fastapi.responses import JSONResponse
 from helpers.log import get_logger

backend/api/endpoints/documents.py

@@ -1,14 +1,14 @@
 from pathlib import Path
 from typing import Annotated
 
-from bot.memory.document_registry import DocumentRegistry
-from bot.memory.vector_database.id_generator import generate_id
-from core.config import settings
-from document_loader.loader import DirectoryLoader
+from config import settings
 from fastapi import APIRouter, File, HTTPException, UploadFile
 from helpers.log import get_logger
-from memory_builder import split_chunks
+from memory.vector_database.id_generator import generate_id
 from schemas.documents import DocumentInfo, DocumentListResponse, DocumentUploadResponse
+from services.ingest_documents_service.document_loader.loader import DirectoryLoader
+from services.ingest_documents_service.document_loader.text_splitter import split_chunks
+from services.ingest_documents_service.document_registry import DocumentRegistry
 
 from api.deps import SessionDep, VectorDatabaseDep
 

backend/api/endpoints/health.py

@@ -1,4 +1,4 @@
-from core.config import settings
+from config import settings
 from fastapi import APIRouter
 from schemas.health import HealthResponse
 

backend/api/services/chat_stream.py

@@ -1,17 +1,17 @@
 import time
 
-from bot.conversation.conversation_handler import (
+from config import settings
+from fastapi import WebSocket
+from helpers.log import get_logger
+from helpers.prettier import prettify_source
+from schemas.chat import ChatRequest
+from services.chat_service.conversation_handler import (
     answer,
     answer_with_context,
     extract_content_after_reasoning,
     refine_question,
 )
-from bot.conversation.ctx_strategy import get_ctx_synthesis_strategy
-from core.config import settings
-from fastapi import WebSocket
-from helpers.log import get_logger
-from helpers.prettier import prettify_source
-from schemas.chat import ChatRequest
+from services.chat_service.ctx_strategy import get_ctx_synthesis_strategy
 
 from api.deps import ChatHistoryDep, LlamaCppClientDep, VectorDatabaseDep