6 views

microsoft/markitdown

Python tool for converting files and office documents to Markdown.

Python★ 181,868+2,047 that day#5 on trending

Analyzed September 9, 2026 at b6e8bbd (main) with jscpd 5.4.0 in 0.1s. Code duplication stayed at 7.68% across 2 appearances since Sep 8.

7.68%duplicated code · 709 of 9,231 lines

39clones in code

53code files of 108 scanned

6.59%duplicated over all files · 91 clones

The code number leaves out tests, fixtures, docs, examples, data files, styles, translations and vendored code.

Largest clones in the code

24 lines in 4 placespython100 tokens

packages/markitdown/src/markitdown/converters/_audio_converter.py:28–51packages/markitdown/src/markitdown/converters/_csv_converter.py:54–74packages/markitdown/src/markitdown/converters/_html_converter.py:24–47packages/markitdown/src/markitdown/converters/_zip_converter.py:69–92

28def accepts(
29    self,
30    file_stream: BinaryIO,
31    stream_info: StreamInfo,
32    **kwargs: Any,  # Options to pass to the converter
33) -> bool:
34    mimetype = (stream_info.mimetype or "").lower()
35    extension = (stream_info.extension or "").lower()
36
37    if extension in ACCEPTED_FILE_EXTENSIONS:

54def accepts(
55    self,
56    file_stream: BinaryIO,
57    stream_info: StreamInfo,
58    **kwargs: Any,  # Options to pass to the converter
59) -> bool:
60    mimetype = (stream_info.mimetype or "").lower()
61    extension = (stream_info.extension or "").lower()
62    if extension in ACCEPTED_FILE_EXTENSIONS:
63        return True

First 10 of 24 lines, the second copy on the right.

31 lines in 3 placespython143 tokens

packages/markitdown/src/markitdown/converters/_docx_converter.py:45–75packages/markitdown/src/markitdown/converters/_epub_converter.py:28–58packages/markitdown/src/markitdown/converters/_pptx_converter.py:36–66

45Converts DOCX files to Markdown. Style information (e.g., headings) and tables are preserved where possible.
46"""
47
48def __init__(self):
49    super().__init__()
50    self._html_converter = HtmlConverter()
51
52def accepts(
53    self,
54    file_stream: BinaryIO,

First 10 of 31 lines.

16 lines in 4 placespython54 tokens
packages/markitdown-ocr/src/markitdown_ocr/_docx_converter_with_ocr.py:70–85 packages/markitdown-ocr/src/markitdown_ocr/_pdf_converter_with_ocr.py:153–168 packages/markitdown-ocr/src/markitdown_ocr/_pptx_converter_with_ocr.py:49–64 packages/markitdown/src/markitdown/converters/_pdf_converter.py:515–530
```
70    ):
71        return True
72
73    return False
74
75def convert(
76    self,
77    file_stream: BinaryIO,
78    stream_info: StreamInfo,
79    **kwargs: Any,
```
First 10 of 16 lines.

7 more

42 lines in 2 placespython152 tokens
packages/markitdown/src/markitdown/converters/_image_converter.py:95–136 packages/markitdown/src/markitdown/converters/_llm_caption.py:9–48
26 lines in 2 placespython148 tokens
packages/markitdown-ocr/src/markitdown_ocr/_pptx_converter_with_ocr.py:155–180 packages/markitdown/src/markitdown/converters/_pptx_converter.py:187–212
18 lines in 3 placespython52 tokens
packages/markitdown/src/markitdown/converters/_bing_serp_converter.py:45–62 packages/markitdown/src/markitdown/converters/_wikipedia_converter.py:39–56 packages/markitdown/src/markitdown/converters/_youtube_converter.py:76–93
23 lines in 2 placespython102 tokens
packages/markitdown/src/markitdown/converters/_csv_converter.py:52–74 packages/markitdown/src/markitdown/converters/_docx_converter.py:50–75
27 lines in 2 placespython101 tokens
packages/markitdown-sample-plugin/src/markitdown_sample_plugin/_plugin.py:36–62 packages/markitdown/src/markitdown/converters/_pdf_converter.py:499–525
26 lines in 2 placespython96 tokens
packages/markitdown-sample-plugin/src/markitdown_sample_plugin/_plugin.py:36–61 packages/markitdown/src/markitdown/converters/_image_converter.py:18–43
18 lines in 2 placespython91 tokens
packages/markitdown-ocr/src/markitdown_ocr/_docx_converter_with_ocr.py:48–65 packages/markitdown-ocr/src/markitdown_ocr/_xlsx_converter_with_ocr.py:30–47

Largest clones over all files, tests and data included

17 lines in 4 placespythontests108 tokens

packages/markitdown/tests/test_pdf_tables.py:735–751packages/markitdown/tests/test_pdf_tables.py:788–804packages/markitdown/tests/test_pdf_tables.py:830–846packages/markitdown/tests/test_pdf_tables.py:913–929

735)
736
737if not os.path.exists(pdf_path):
738    pytest.skip(f"Test file not found: {pdf_path}")
739
740if not os.path.exists(expected_path):
741    pytest.skip(f"Expected output not found: {expected_path}")
742
743result = markitdown.convert(pdf_path)
744actual_output = result.text_content

First 10 of 17 lines.

24 lines in 4 placespython100 tokens

28def accepts(
29    self,
30    file_stream: BinaryIO,
31    stream_info: StreamInfo,
32    **kwargs: Any,  # Options to pass to the converter
33) -> bool:
34    mimetype = (stream_info.mimetype or "").lower()
35    extension = (stream_info.extension or "").lower()
36
37    if extension in ACCEPTED_FILE_EXTENSIONS:

54def accepts(
55    self,
56    file_stream: BinaryIO,
57    stream_info: StreamInfo,
58    **kwargs: Any,  # Options to pass to the converter
59) -> bool:
60    mimetype = (stream_info.mimetype or "").lower()
61    extension = (stream_info.extension or "").lower()
62    if extension in ACCEPTED_FILE_EXTENSIONS:
63        return True

First 10 of 24 lines, the second copy on the right.

31 lines in 3 placespython143 tokens

45Converts DOCX files to Markdown. Style information (e.g., headings) and tables are preserved where possible.
46"""
47
48def __init__(self):
49    super().__init__()
50    self._html_converter = HtmlConverter()
51
52def accepts(
53    self,
54    file_stream: BinaryIO,

First 10 of 31 lines.

8 lines in 4 placespythontests91 tokens
packages/markitdown/tests/test_docx_math.py:8–15 packages/markitdown/tests/test_docx_math.py:22–29 packages/markitdown/tests/test_docx_math.py:38–45 packages/markitdown/tests/test_docx_math.py:52–59
7 lines in 4 placesmarkdowndocs and examples72 tokens
README.md:398–404 packages/markitdown-mcp/README.md:136–142 packages/markitdown-sample-plugin/README.md:105–111 packages/markitdown/README.md:49–55

By format, code only

Format	Files	Lines	Clones	Duplicated lines	Duplication
python	43	8,413	39	709	8.43%
bash	9	644	0	0	0%
perl	1	174	0	0	0%

By format, all files

Format	Files	Lines	Clones	Duplicated lines	Duplication
python	70	16,006	83	1,319	8.24%
markdown	14	1,545	5	36	2.33%
toml	4	311	2	36	11.58%
yaml	2	66	1	12	18.18%
markup	3	2,448	0	0	0%
bash	12	688	0	0	0%
perl	1	174	0	0	0%
json	2	60	0	0	0%

Trending appearances

Day	Rank	Stars	Files	Lines	Clones	All files	Code	Health	Commit
Sep 9, 2026	#5	181,868 (+2,047)	108	21,298	91	6.59%	7.68%	—	`b6e8bbd`
Sep 8, 2026	#1	180,696 (+886)	108	21,298	91	6.59%	7.68%	—	`b6e8bbd`

Want this for your own project? Install jscpd and run jscpd ., or measure it the way the baselines were measured.