add Markdown as a support RAG mime type

langchain-ai · May 25, 2024 · cdd9459 · cdd9459
1 parent b85eb8d
commit cdd9459
Show file tree

Hide file tree

Showing 4 changed files with 11 additions and 1 deletion.
diff --git a/backend/app/parsing.py b/backend/app/parsing.py
@@ -1,12 +1,13 @@
 """Module contains logic for parsing binary blobs into text."""
-from langchain_community.document_loaders.parsers import BS4HTMLParser, PDFMinerParser
+from langchain_community.document_loaders.parsers import BS4HTMLParser, PDFMinerParser, MarkdownParser
 from langchain_community.document_loaders.parsers.generic import MimeTypeBasedParser
 from langchain_community.document_loaders.parsers.msword import MsWordParser
 from langchain_community.document_loaders.parsers.txt import TextParser
 
 HANDLERS = {
     "application/pdf": PDFMinerParser(),
     "text/plain": TextParser(),
+    "text/markdown": MarkdownParser(),
     "text/html": BS4HTMLParser(),
     "application/msword": MsWordParser(),
     "application/vnd.openxmlformats-officedocument.wordprocessingml.document": (

diff --git a/backend/tests/unit_tests/agent_executor/test_parsing.py b/backend/tests/unit_tests/agent_executor/test_parsing.py
@@ -15,6 +15,7 @@ def test_list_of_supported_mimetypes() -> None:
         "application/vnd.openxmlformats-officedocument.wordprocessingml.document",
         "text/html",
         "text/plain",
+        "text/markdown",
     ]
 
 

diff --git a/backend/tests/unit_tests/agent_executor/test_upload.py b/backend/tests/unit_tests/agent_executor/test_upload.py
@@ -46,4 +46,5 @@ def test_mimetype_guessing() -> None:
         "sample.pdf": "application/pdf",
         "sample.rtf": "application/rtf",
         "sample.txt": "text/plain",
+        "sample.md": "text/markdown",
     } == name_to_mime
diff --git a/backend/tests/unit_tests/fixtures/sample.md b/backend/tests/unit_tests/fixtures/sample.md
@@ -0,0 +1,7 @@
+# 🦜️ LangChain
+
+## Heading 2
+
+### Heading 3
+
+#### Heading 4