luciano/markitdown
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

add xls converter with tests

Browse source
This commit is contained in:
yeungadrian 2024-12-20 12:50:08 +00:00
parent 5f58dd89f0
commit 488f89353d
3 changed files with 37 additions and 1 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

26
src/markitdown/_markitdown.py
View file

@ -723,7 +723,30 @@ class XlsxConverter(HtmlConverter):
if extension.lower() != ".xlsx": if extension.lower() != ".xlsx":
return None return None
sheets = pd.read_excel(local_path, sheet_name=None) sheets = pd.read_excel(local_path, sheet_name=None, engine="openpyxl")
md_content = ""
for s in sheets:
md_content += f"## {s}\n"
html_content = sheets[s].to_html(index=False)
md_content += self._convert(html_content).text_content.strip() + "\n\n"
return DocumentConverterResult(
title=None,
text_content=md_content.strip(),
)
class XlsConverter(HtmlConverter):
"""
Converts XLS files to Markdown, with each sheet presented as a separate Markdown table.
"""
def convert(self, local_path, **kwargs) -> Union[None, DocumentConverterResult]:
# Bail if not a XLS
extension = kwargs.get("file_extension", "")
if extension.lower() != ".xls":
return None
sheets = pd.read_excel(local_path, sheet_name=None, engine="xlrd")
md_content = "" md_content = ""
for s in sheets: for s in sheets:
md_content += f"## {s}\n" md_content += f"## {s}\n"
@ -1277,6 +1300,7 @@ class MarkItDown:
self.register_page_converter(BingSerpConverter()) self.register_page_converter(BingSerpConverter())
self.register_page_converter(DocxConverter()) self.register_page_converter(DocxConverter())
self.register_page_converter(XlsxConverter()) self.register_page_converter(XlsxConverter())
self.register_page_converter(XlsConverter())
self.register_page_converter(PptxConverter()) self.register_page_converter(PptxConverter())
self.register_page_converter(WavConverter()) self.register_page_converter(WavConverter())
self.register_page_converter(Mp3Converter()) self.register_page_converter(Mp3Converter())

BIN
tests/test_files/test.xls vendored Normal file
View file

Binary file not shown.

12
tests/test_markitdown.py
View file

@ -54,6 +54,12 @@ XLSX_TEST_STRINGS = [
"affc7dad-52dc-4b98-9b5d-51e65d8a8ad0", "affc7dad-52dc-4b98-9b5d-51e65d8a8ad0",
] ]
XLS_TEST_STRINGS = [
"## 09060124-b5e7-4717-9d07-3c046eb",
"6ff4173b-42a5-4784-9b19-f49caff4d93d",
"affc7dad-52dc-4b98-9b5d-51e65d8a8ad0",
]
DOCX_TEST_STRINGS = [ DOCX_TEST_STRINGS = [
"314b0a30-5b04-470b-b9f7-eed2c2bec74a", "314b0a30-5b04-470b-b9f7-eed2c2bec74a",
"49e168b7-d2ae-407f-a055-2167576f39a1", "49e168b7-d2ae-407f-a055-2167576f39a1",
@ -167,6 +173,12 @@ 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 XLS processing
result = markitdown.convert(os.path.join(TEST_FILES_DIR, "test.xls"))
for test_string in XLS_TEST_STRINGS:
text_content = result.text_content.replace("\\", "")
assert test_string in text_content
# Test DOCX processing # Test DOCX processing
result = markitdown.convert(os.path.join(TEST_FILES_DIR, "test.docx")) result = markitdown.convert(os.path.join(TEST_FILES_DIR, "test.docx"))
for test_string in DOCX_TEST_STRINGS: for test_string in DOCX_TEST_STRINGS: