From 87727b3d1ea62610f3848eb752c335b0cfaa8bef Mon Sep 17 00:00:00 2001
From: Falko Habel <falko.habel@gmx.de>
Date: Mon, 20 May 2024 12:27:03 +0200
Subject: [PATCH] improved error handling

---
 requirements.txt |  3 ++-
 scripts/Rag.py   | 13 ++++++++-----
 2 files changed, 10 insertions(+), 6 deletions(-)

diff --git a/requirements.txt b/requirements.txt
index e06b397..3cabd84 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -6,4 +6,5 @@ langchain_community
 langchain
 pathlib
 unstructured
-markdown
\ No newline at end of file
+markdown
+requests
\ No newline at end of file
diff --git a/scripts/Rag.py b/scripts/Rag.py
index 0c5ee16..4b34789 100644
--- a/scripts/Rag.py
+++ b/scripts/Rag.py
@@ -6,6 +6,7 @@ from langchain_community.embeddings import OllamaEmbeddings
 from langchain_community.vectorstores import Chroma
 from langchain_community.chat_models import ChatOllama
 from langchain.chains import RetrievalQA
+import requests
 
 
 
@@ -40,9 +41,12 @@ class Rag:
     def get_file(self, file_path):
         # Check if the file path starts with 'https://'
         if file_path.startswith('https://'):
-            loader = WebBaseLoader(file_path)
-            data = loader.load()
-            if data is None:
+            try:
+                loader = WebBaseLoader(file_path)
+                data = loader.load()
+                if data is None:
+                    return False
+            except requests.exceptions.SSLError:
                 return False
         else:
             file_type = file_path.split(".")[-1]
@@ -61,8 +65,7 @@ class Rag:
                         loader = PyPDFLoader(file_path=file_path)
                         data = loader.load_and_split()
                     case _:
-                        loader = WebBaseLoader(file_path)
-                        data = loader.load()
+                        return False
             except OSError:
                 return False