From aa100011da73e662dcc6ba87569d70466df185df Mon Sep 17 00:00:00 2001
From: Nikos Papailiou <115017354+NikolaosPapailiou@users.noreply.github.com>
Date: Wed, 25 Sep 2024 09:28:26 +0300
Subject: [PATCH] Add explicit markdown parser (#532)

---
 .../tiledb/vector_search/object_readers/directory_reader.py    | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/apis/python/src/tiledb/vector_search/object_readers/directory_reader.py b/apis/python/src/tiledb/vector_search/object_readers/directory_reader.py
index 1a815468d..2e861d262 100644
--- a/apis/python/src/tiledb/vector_search/object_readers/directory_reader.py
+++ b/apis/python/src/tiledb/vector_search/object_readers/directory_reader.py
@@ -355,6 +355,7 @@ def __init__(
             handlers={
                 "application/pdf": PyMuPDFParser(),
                 "text/plain": TextParser(),
+                "text/markdown": TextParser(),
                 "text/html": BS4HTMLParser(),
                 "application/msword": MsWordParser(),
                 "application/vnd.openxmlformats-officedocument.wordprocessingml.document": (
@@ -385,7 +386,7 @@ def lazy_load(
             mime_type = mimetypes.guess_type(self.uri)[0]
             f = vfs.open(self.uri)
 
-        if mime_type is None:
+        if mime_type is None or mime_type.startswith("text"):
             mime_type = "text/plain"
 
         if mime_type.startswith("image/"):