Merge pull request 'general_fixes' (#25) from general_fixes into main

Reviewed-on: AI_team/generic-RAG-demo#25

generic_rag/app.py

@@ -5,15 +5,16 @@ import os
 from pathlib import Path
 
 import chainlit as cl
-from backend.models import ChatBackend, EmbeddingBackend, get_chat_model, get_embedding_model
 from chainlit.cli import run_chainlit
-from graphs.cond_ret_gen import CondRetGenLangGraph
-from graphs.ret_gen import RetGenLangGraph
 from langchain_chroma import Chroma
-from parsers.parser import add_pdf_files, add_urls
 
-logging.basicConfig(level=logging.DEBUG)
-logger = logging.getLogger(__name__)
+from generic_rag.backend.models import ChatBackend, EmbeddingBackend, get_chat_model, get_embedding_model
+from generic_rag.graphs.cond_ret_gen import CondRetGenLangGraph
+from generic_rag.graphs.ret_gen import RetGenLangGraph
+from generic_rag.parsers.parser import add_pdf_files, add_urls
+
+logger = logging.getLogger("sogeti-rag")
+logger.setLevel(logging.DEBUG)
 
 system_prompt = (
     "You are an assistant for question-answering tasks. "
@@ -126,12 +127,9 @@ async def add_sources(chainlit_response: cl.Message, pdf_sources: dict, web_sour
     if len(pdf_sources) > 0:
         await chainlit_response.stream_token("\n\nThe following PDF source were consulted:\n")
         for source, page_numbers in pdf_sources.items():
-            page_numbers = list(page_numbers)
-            page_numbers.sort()
-            # display="side" seems to be not supported by chainlit for PDF's, so we use "inline" instead.
-            chainlit_response.elements.append(cl.Pdf(name="pdf", display="inline", path=source, page=page_numbers[0]))
-            await chainlit_response.update()
-            await chainlit_response.stream_token(f"- '{source}' on page(s): {page_numbers}\n")
+            filename = Path(source).name
+            await chainlit_response.stream_token(f"- {filename} on page(s): {sorted(page_numbers)}\n")
+            chainlit_response.elements.append(cl.Pdf(name=filename, display="side", path=source, page=sorted(page_numbers)[0]))
 
     if len(web_sources) > 0:
         await chainlit_response.stream_token("\n\nThe following web sources were consulted:\n")

generic_rag/graphs/cond_ret_gen.py

@@ -16,7 +16,7 @@ from langgraph.graph import END, MessagesState, StateGraph
 from langgraph.prebuilt import InjectedStore, ToolNode, tools_condition
 from typing_extensions import Annotated
 
-logger = logging.getLogger(__name__)
+logger = logging.getLogger("sogeti-rag")
 
 class CondRetGenLangGraph:
     def __init__(

generic_rag/graphs/ret_gen.py

@@ -12,7 +12,7 @@ from langgraph.checkpoint.memory import MemorySaver
 from langgraph.graph import END, START, StateGraph
 from typing_extensions import List, TypedDict
 
-logger = logging.getLogger(__name__)
+logger = logging.getLogger("sogeti-rag")
 
 
 class State(TypedDict):