tdpeuter/open-webui
tdpeuter/open-webui
1
Fork 0
forked from open-webui/open-webui
Code Issues Pull requests 1 Projects 1 Releases 2 Packages 1 Activity Actions

include html langchain loader for RAG

Browse source
This commit is contained in:
Doug Danat 2024-03-25 09:50:53 +01:00
parent eb004ccfc2
commit 784a6ec85e
1 changed files with 3 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

3
backend/apps/rag/main.py
View file

@ -21,6 +21,7 @@ from langchain_community.document_loaders import (
TextLoader, TextLoader,
PyPDFLoader, PyPDFLoader,
CSVLoader, CSVLoader,
UnstructuredHTMLLoader,
Docx2txtLoader, Docx2txtLoader,
UnstructuredEPubLoader, UnstructuredEPubLoader,
UnstructuredWordDocumentLoader, UnstructuredWordDocumentLoader,
@ -402,6 +403,8 @@ def get_loader(filename: str, file_content_type: str, file_path: str):
loader = UnstructuredRSTLoader(file_path, mode="elements") loader = UnstructuredRSTLoader(file_path, mode="elements")
elif file_ext == "xml": elif file_ext == "xml":
loader = UnstructuredXMLLoader(file_path) loader = UnstructuredXMLLoader(file_path)
elif file_ext in ["htm", "html"]:
loader = UnstructuredHTMLLoader(file_path)
elif file_ext == "md": elif file_ext == "md":
loader = UnstructuredMarkdownLoader(file_path) loader = UnstructuredMarkdownLoader(file_path)
elif file_content_type == "application/epub+zip": elif file_content_type == "application/epub+zip":