From 6efcce459de3e28e4fb87a303aa0ab12fe0ca290 Mon Sep 17 00:00:00 2001
From: Manohar Paturi <186662190+ManoharPaturi@users.noreply.github.com>
Date: Fri, 18 Sep 2026 16:33:41 +0530
Subject: [PATCH 1/2] fix(rss): emit item and entry links as heading anchors
RSS items and Atom entries carry a (or Atom ), but the
converter never read it: the markdown ended up with bare titles and no
way to reach the post. Emit the title as a markdown link to the resolved
URL instead (relative hrefs resolve against the feed URL).
Signed-off-by: Manohar Paturi <186662190+ManoharPaturi@users.noreply.github.com>
---
.../markitdown/converters/_rss_converter.py | 25 +++++++++++++++++--
1 file changed, 23 insertions(+), 2 deletions(-)
diff --git a/packages/markitdown/src/markitdown/converters/_rss_converter.py b/packages/markitdown/src/markitdown/converters/_rss_converter.py
index ce768b8473..5389def46f 100644
--- a/packages/markitdown/src/markitdown/converters/_rss_converter.py
+++ b/packages/markitdown/src/markitdown/converters/_rss_converter.py
@@ -180,9 +180,13 @@ def _parse_atom_type(
entry_summary, summary_is_markup = self._get_atom_content(entry, "summary")
entry_updated = self._get_flattened_text(entry, "updated")
entry_content, content_is_markup = self._get_atom_content(entry, "content")
+ entry_link = self._get_entry_link(entry, base_url=doc.documentURI or "")
if entry_title:
- md_text += f"\n## {entry_title}\n"
+ heading = f"## {entry_title}"
+ if entry_link:
+ heading = f"[{entry_title}]({entry_link})"
+ md_text += f"\n{heading}\n"
if entry_updated:
md_text += f"Updated on: {entry_updated}\n"
body_parts = (
@@ -208,6 +212,19 @@ def _parse_atom_type(
title=title,
)
+ def _get_entry_link(self, entry: Element, *, base_url: str) -> str | None:
+ """Return the entry's own URL: an Atom , or an RSS 's text."""
+ if entry.namespaceURI == ATOM_NAMESPACE:
+ node = self._get_child(entry, "link")
+ if node is None:
+ return None
+ href = (node.getAttribute("href") or "").strip()
+ return _resolve_url(base_url, href) if href else None
+ text = self._get_flattened_text(entry, "link")
+ if not text:
+ return None
+ return _resolve_url(base_url, text.strip())
+
def _get_atom_content(
self, entry: Element, tag_name: str
) -> tuple[Union[str, None], bool]:
@@ -308,9 +325,13 @@ def _parse_rss_type(
description = self._get_data_by_tag_name(item, "description", kind="html")
pubDate = self._get_flattened_text(item, "pubDate")
content = self._get_data_by_tag_name(item, "content:encoded", kind="html")
+ item_link = self._get_entry_link(item, base_url=doc.documentURI or "")
if title:
- md_text += f"\n## {title}\n"
+ heading = f"## {title}"
+ if item_link:
+ heading = f"[{title}]({item_link})"
+ md_text += f"\n{heading}\n"
if pubDate:
md_text += f"Published on: {pubDate}\n"
body_parts = (
From 40d6912b839c302ae5c93a7bd30d635896ffce10 Mon Sep 17 00:00:00 2001
From: Manohar Paturi <186662190+ManoharPaturi@users.noreply.github.com>
Date: Sat, 19 Sep 2026 23:07:55 +0530
Subject: [PATCH 2/2] fix(rss): keep the heading marker when an item link is
present
Signed-off-by: Manohar Paturi <186662190+ManoharPaturi@users.noreply.github.com>
---
packages/markitdown/src/markitdown/converters/_rss_converter.py | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/packages/markitdown/src/markitdown/converters/_rss_converter.py b/packages/markitdown/src/markitdown/converters/_rss_converter.py
index 5389def46f..b234fe4cdc 100644
--- a/packages/markitdown/src/markitdown/converters/_rss_converter.py
+++ b/packages/markitdown/src/markitdown/converters/_rss_converter.py
@@ -185,7 +185,7 @@ def _parse_atom_type(
if entry_title:
heading = f"## {entry_title}"
if entry_link:
- heading = f"[{entry_title}]({entry_link})"
+ heading = f"## [{entry_title}]({entry_link})"
md_text += f"\n{heading}\n"
if entry_updated:
md_text += f"Updated on: {entry_updated}\n"