Commit 784a6ec8 authored by Doug Danat's avatar Doug Danat
Browse files

include html langchain loader for RAG

parent eb004ccf
...@@ -21,6 +21,7 @@ from langchain_community.document_loaders import ( ...@@ -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): ...@@ -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":
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment