Fixed issue #17397 by updating the markdown format in MarkdownReader (#…

…17429)
run-llama · Jan 10, 2025 · 992ef61 · 992ef61
1 parent 652e37a
commit 992ef61
Show file tree

Hide file tree

Showing 2 changed files with 4 additions and 4 deletions.
diff --git a/...x-integrations/readers/llama-index-readers-file/llama_index/readers/file/markdown/base.py b/...x-integrations/readers/llama-index-readers-file/llama_index/readers/file/markdown/base.py
@@ -75,9 +75,9 @@ def markdown_to_tups(self, markdown_text: str) -> List[Tuple[Optional[str], str]
         ]
 
     def remove_images(self, content: str) -> str:
-        """Remove images in markdown content."""
-        pattern = r"!{1}\[\[(.*)\]\]"
-        return re.sub(pattern, "", content)
+        """Remove images in markdown content but keep the description."""
+        pattern = r"![(.?)](.?)"
+        return re.sub(pattern, r"\1", content)
 
     def remove_hyperlinks(self, content: str) -> str:
         """Remove hyperlinks in markdown content."""

diff --git a/llama-index-integrations/readers/llama-index-readers-file/pyproject.toml b/llama-index-integrations/readers/llama-index-readers-file/pyproject.toml
@@ -51,7 +51,7 @@ license = "MIT"
 maintainers = ["FarisHijazi", "Haowjy", "ephe-meral", "hursh-desai", "iamarunbrahma", "jon-chuang", "mmaatouk", "ravi03071991", "sangwongenip", "thejessezhang"]
 name = "llama-index-readers-file"
 readme = "README.md"
-version = "0.4.2"
+version = "0.4.3"
 
 [tool.poetry.dependencies]
 python = ">=3.9,<4.0"