Merge pull request #701 from GSK-FS/issue-47-ingest-html-documents

HTML ingest
PromtEngineer · Jan 6, 2024 · e8d294e · e8d294e
2 parents 0b4fa33 + 7bc14c8
commit e8d294e
Showing 1 changed file with 2 additions and 0 deletions.
diff --git a/constants.py b/constants.py
@@ -6,6 +6,7 @@
 # https://python.langchain.com/en/latest/modules/indexes/document_loaders/examples/excel.html?highlight=xlsx#microsoft-excel
 from langchain.document_loaders import CSVLoader, PDFMinerLoader, TextLoader, UnstructuredExcelLoader, Docx2txtLoader
 from langchain.document_loaders import UnstructuredFileLoader, UnstructuredMarkdownLoader
+from langchain.document_loaders import UnstructuredHTMLLoader
 
 
 # load_dotenv()
@@ -43,6 +44,7 @@
 
 # https://python.langchain.com/en/latest/_modules/langchain/document_loaders/excel.html#UnstructuredExcelLoader
 DOCUMENT_MAP = {
+ ".html": UnstructuredHTMLLoader,
  ".txt": TextLoader,
  ".md": UnstructuredMarkdownLoader,
  ".py": TextLoader,