Update helpers.py

helpers.py

@@ -269,13 +269,12 @@ def get_web_documents(base_urls=['https://nct.neu.edu.vn/']):
 def load_text_data(file_path):
     """Load text content from a DOCX file (tables removed)."""
     text_splitter = RecursiveCharacterTextSplitter(chunk_size=10000, chunk_overlap=2500)
-    loader = DoclingLoader(
-        file_path=file_path,
-        export_type=ExportType.MARKDOWN,  # Enable internal chunking,
-        chunker = text_splitter,
-        # convert_kwargs={"input_format": "docx"}  # Specify the input format
-    )
-    docs =  loader.load()
+    # loader = DoclingLoader(
+    #     file_path=file_path,
+    #     export_type=ExportType.MARKDOWN,  # Enable internal chunking,
+    #     # chunker = text_splitter
+    # )
+    loader = UnstructuredWordDocumentLoader(file_path)
     chunks = text_splitter.split_documents(docs)
     # You can wrap each chunk back into a Document if needed.
     return chunks