Merge pull request 'Small fixes' (#22) from small_fixes into main

Reviewed-on: AI_team/generic-RAG-demo#22

generic_rag/graphs/cond_ret_gen.py

@@ -8,7 +8,7 @@ from langchain_chroma import Chroma
 from langchain_core.documents import Document
 from langchain_core.embeddings import Embeddings
 from langchain_core.language_models.chat_models import BaseChatModel
-from langchain_core.messages import BaseMessage, HumanMessage, SystemMessage
+from langchain_core.messages import BaseMessage, SystemMessage
 from langchain_core.tools import tool
 from langchain_core.runnables.config import RunnableConfig
 from langgraph.checkpoint.memory import MemorySaver

generic_rag/graphs/ret_gen.py

@@ -61,7 +61,6 @@ class RetGenLangGraph:
             return pdf_sources
 
         for doc in self.last_retrieved_docs:
-            source_candidate = doc.metadata["source"]
             if "source" in doc.metadata and Path(doc.metadata["source"]).suffix.lower() == ".pdf":
                 source = doc.metadata["source"]
             else:
@@ -71,19 +70,11 @@ class RetGenLangGraph:
                 pdf_sources[source] = set()
 
             # The page numbers are in the `page_numer` and `page` fields.
-            try:
-                page_number = doc.metadata["page_number"]
-            except KeyError:
-                pass
-            else:
-                pdf_sources[source].add(page_number)
+            if "page_number" in doc.metadata:
+                pdf_sources[source].add(doc.metadata["page_number"])
 
-            try:
-                page_number = doc.metadata["page"]
-            except KeyError:
-                pass
-            else:
-                pdf_sources[source].add(page_number)
+            if "page" in doc.metadata:
+                pdf_sources[source].add(doc.metadata["page"])
 
             if len(pdf_sources[source]) == 0:
                 logging.warning(f"PDF source {source} has no page number. Please check the metadata of the document.")
@@ -100,10 +91,7 @@ class RetGenLangGraph:
             return web_sources
 
         for doc in self.last_retrieved_docs:
-            try:
-                if doc.metadata["filetype"] == "web":
-                    web_sources.add(doc.metadata["source"])
-            except KeyError:
-                continue
+            if "filetype" in doc.metadata and doc.metadata["filetype"] == "web":
+                web_sources.add(doc.metadata["source"])
 
         return web_sources