From b39c3fb21026ef5de1195939bb1f817bc1a2972e Mon Sep 17 00:00:00 2001
From: Bruce-anle <840596168@qq.com>
Date: Sat, 9 May 2026 00:52:11 +0800
Subject: [PATCH] fix: convert markdown headers and footers

Background: convert_to_markdown reads word/header*.xml and word/footer*.xml, but passed w:hdr/w:ftr roots to parse_body_to_markdown. That parser only looked for w:body, so header/footer content was skipped.\n\nChanges: allow parse_body_to_markdown to traverse w:hdr and w:ftr roots directly while preserving normal w:document/w:body behavior.\n\nVerification: /home/brucean/doc4agent/.venv/bin/python -m pytest tests -q -p no:cacheprovider passed.
---
 .../converters/markdown_converter.py          |  3 ++
 tests/test_markdown_header_footer.py          | 48 +++++++++++++++++++
 2 files changed, 51 insertions(+)
 create mode 100644 tests/test_markdown_header_footer.py

diff --git a/docx2everything/converters/markdown_converter.py b/docx2everything/converters/markdown_converter.py
index 0ff408f..e7c0842 100644
--- a/docx2everything/converters/markdown_converter.py
+++ b/docx2everything/converters/markdown_converter.py
@@ -511,6 +511,9 @@ def parse_body_to_markdown(root, numbering_info=None, hyperlinks=None, images=No
     markdown_parts = []
     body = root.find(qn('w:body'))
     
+    if body is None and root.tag in (qn('w:hdr'), qn('w:ftr')):
+        body = root
+
     if body is None:
         return ''
     
diff --git a/tests/test_markdown_header_footer.py b/tests/test_markdown_header_footer.py
new file mode 100644
index 0000000..43d016d
--- /dev/null
+++ b/tests/test_markdown_header_footer.py
@@ -0,0 +1,48 @@
+import xml.etree.ElementTree as ET
+
+from docx2everything.converters.markdown_converter import parse_body_to_markdown
+
+
+W_NS = "http://schemas.openxmlformats.org/wordprocessingml/2006/main"
+
+
+def xml_root(xml):
+    return ET.fromstring(xml)
+
+
+def test_header_root_is_converted_to_markdown():
+    root = xml_root(f"""
+    <w:hdr xmlns:w="{W_NS}">
+      <w:p><w:r><w:t>Header text</w:t></w:r></w:p>
+    </w:hdr>
+    """)
+
+    markdown = parse_body_to_markdown(root)
+
+    assert markdown == "Header text"
+
+
+def test_footer_root_is_converted_to_markdown():
+    root = xml_root(f"""
+    <w:ftr xmlns:w="{W_NS}">
+      <w:p><w:r><w:t>Footer text</w:t></w:r></w:p>
+    </w:ftr>
+    """)
+
+    markdown = parse_body_to_markdown(root)
+
+    assert markdown == "Footer text"
+
+
+def test_document_body_root_still_converts_to_markdown():
+    root = xml_root(f"""
+    <w:document xmlns:w="{W_NS}">
+      <w:body>
+        <w:p><w:r><w:t>Body text</w:t></w:r></w:p>
+      </w:body>
+    </w:document>
+    """)
+
+    markdown = parse_body_to_markdown(root)
+
+    assert markdown == "Body text"