orhan/test
1
1
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

pass 'style_map' kwarg to mammoth when converting docx

Browse Source
This commit is contained in:
Ville Puuska
2024-12-15 16:59:21 +02:00
parent 81e3f24acd
commit 0704b0b6ff
3 changed files with 23 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
src/markitdown/_markitdown.py
View File

@@ -492,7 +492,9 @@ class DocxConverter(HtmlConverter):
result = None result = None
with open(local_path, "rb") as docx_file: with open(local_path, "rb") as docx_file:
result = mammoth.convert_to_html(docx_file) style_map = kwargs.get("style_map", None)
result = mammoth.convert_to_html(docx_file, style_map=style_map)
html_content = result.value html_content = result.value
result = self._convert(html_content) result = self._convert(html_content)

BIN
tests/test_files/test_with_comment.docx Executable file
View File

Binary file not shown.

20
tests/test_markitdown.py
View File

@@ -51,6 +51,17 @@ DOCX_TEST_STRINGS = [
"AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversation", "AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversation",
] ]
DOCX_COMMENT_TEST_STRINGS = [
"314b0a30-5b04-470b-b9f7-eed2c2bec74a",
"49e168b7-d2ae-407f-a055-2167576f39a1",
"## d666f1f7-46cb-42bd-9a39-9a39cf2a509f",
"# Abstract",
"# Introduction",
"AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversation",
"This is a test comment. 12df-321a",
"Yet another comment in the doc. 55yiyi-asd09",
]
PPTX_TEST_STRINGS = [ PPTX_TEST_STRINGS = [
"2cdda5c8-e50e-4db4-b5f0-9722a649f455", "2cdda5c8-e50e-4db4-b5f0-9722a649f455",
"04191ea8-5c73-4215-a1d3-1cfb43aaaf12", "04191ea8-5c73-4215-a1d3-1cfb43aaaf12",
@@ -130,6 +141,15 @@ def test_markitdown_local() -> None:
text_content = result.text_content.replace("\\", "") text_content = result.text_content.replace("\\", "")
assert test_string in text_content assert test_string in text_content
# Test DOCX processing, with comments
result = markitdown.convert(
os.path.join(TEST_FILES_DIR, "test_with_comment.docx"),
style_map="comment-reference => ",
)
for test_string in DOCX_COMMENT_TEST_STRINGS:
text_content = result.text_content.replace("\\", "")
assert test_string in text_content
# Test PPTX processing # Test PPTX processing
result = markitdown.convert(os.path.join(TEST_FILES_DIR, "test.pptx")) result = markitdown.convert(os.path.join(TEST_FILES_DIR, "test.pptx"))
for test_string in PPTX_TEST_STRINGS: for test_string in PPTX_TEST_STRINGS: