From 089a63e0c68bb1c2693a15a8469a278dc358b111 Mon Sep 17 00:00:00 2001
From: Jannik Streidl <jannik.stdl@gmail.com>
Date: Tue, 5 Mar 2024 22:25:25 +0100
Subject: [PATCH] feat: added ocr functionality to the pdf loader

---
 backend/apps/rag/main.py | 2 +-
 backend/requirements.txt | 1 +
 2 files changed, 2 insertions(+), 1 deletion(-)

diff --git a/backend/apps/rag/main.py b/backend/apps/rag/main.py
index 2a8b2a49..ee07d51a 100644
--- a/backend/apps/rag/main.py
+++ b/backend/apps/rag/main.py
@@ -419,7 +419,7 @@ def get_loader(filename: str, file_content_type: str, file_path: str):
     ]
 
     if file_ext == "pdf":
-        loader = PyPDFLoader(file_path)
+        loader = PyPDFLoader(file_path, extract_images=True)
     elif file_ext == "csv":
         loader = CSVLoader(file_path)
     elif file_ext == "rst":
diff --git a/backend/requirements.txt b/backend/requirements.txt
index 0cacacd8..9de3df96 100644
--- a/backend/requirements.txt
+++ b/backend/requirements.txt
@@ -33,6 +33,7 @@ pandas
 openpyxl
 pyxlsb
 xlrd
+rapidocr-onnxruntime
 
 faster-whisper