Update test_backend_msexcel.py

Signed-off-by: ShiroYasha18 <85089952+ShiroYasha18@users.noreply.github.com>
This commit is contained in:
ShiroYasha18 2025-05-27 02:31:43 +05:30 committed by GitHub
parent 96377cb81e
commit 1f645e62c8
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194

View File

@ -16,13 +16,13 @@ _log = logging.getLogger(__name__)
GENERATE = GEN_TEST_DATA
def get_xlsx_paths():
def get_excel_paths():
# Define the directory you want to search
directory = Path("./tests/data/xlsx/")
# List all PDF files in the directory and its subdirectories
pdf_files = sorted(directory.rglob("*.xlsx"))
return pdf_files
# List all Excel files in the directory and its subdirectories
excel_files = sorted(directory.rglob("*.xlsx")) + sorted(directory.rglob("*.xlsm"))
return excel_files
def get_converter():
@ -35,17 +35,17 @@ def get_converter():
def documents() -> list[tuple[Path, DoclingDocument]]:
documents: list[dict[Path, DoclingDocument]] = []
xlsx_paths = get_xlsx_paths()
excel_paths = get_excel_paths()
converter = get_converter()
for xlsx_path in xlsx_paths:
_log.debug(f"converting {xlsx_path}")
for excel_path in excel_paths:
_log.debug(f"converting {excel_path}")
gt_path = (
xlsx_path.parent.parent / "groundtruth" / "docling_v2" / xlsx_path.name
excel_path.parent.parent / "groundtruth" / "docling_v2" / excel_path.name
)
conv_result: ConversionResult = converter.convert(xlsx_path)
conv_result: ConversionResult = converter.convert(excel_path)
doc: DoclingDocument = conv_result.document
@ -55,7 +55,7 @@ def documents() -> list[tuple[Path, DoclingDocument]]:
return documents
def test_e2e_xlsx_conversions(documents) -> None:
def test_e2e_excel_conversions(documents) -> None:
for gt_path, doc in documents:
pred_md: str = doc.export_to_markdown()
assert verify_export(pred_md, str(gt_path) + ".md"), "export to md"
@ -79,21 +79,31 @@ def test_pages(documents) -> None:
documents: The paths and converted documents.
"""
# number of pages from the backend method
path = next(item for item in get_xlsx_paths() if item.stem == "test-01")
in_doc = InputDocument(
path_or_stream=path,
format=InputFormat.XLSX,
filename=path.stem,
backend=MsExcelDocumentBackend,
)
backend = MsExcelDocumentBackend(in_doc=in_doc, path_or_stream=path)
assert backend.page_count() == 3
# Logic to handle multiple files
file_stems = [ "sample_sales_data", "test-01"]
for stem in file_stems:
path = next(item for item in get_excel_paths() if item.stem == stem)
in_doc = InputDocument(
path_or_stream=path,
format=InputFormat.XLSX,
filename=path.stem,
backend=MsExcelDocumentBackend,
)
backend = MsExcelDocumentBackend(in_doc=in_doc, path_or_stream=path)
# Update the expected page count based on actual content
expected_page_count = 3 # Adjust this value based on the actual number of worksheets this needs to be adjusted for each xlsm and xlsx files independently
assert backend.page_count() == expected_page_count
# number of pages from the converted document
doc = next(item for path, item in documents if path.stem == stem)
assert len(doc.pages) == 3
# page sizes as number of cells
# number of pages from the converted document
doc = next(item for path, item in documents if path.stem == "test-01")
assert len(doc.pages) == 3
# page sizes as number of cells
assert doc.pages.get(1).size.as_tuple() == (3.0, 7.0)
assert doc.pages.get(2).size.as_tuple() == (9.0, 18.0)
assert doc.pages.get(3).size.as_tuple() == (13.0, 36.0)
# for xlsm file just adjust this wrt the xlsm files for test xlsm enable this:
# assert doc.pages.get(1).size.as_tuple() == (4.0, 21.0)
# for xlsx file:
assert doc.pages.get(1).size.as_tuple() == (3.0, 7.0)
assert doc.pages.get(2).size.as_tuple() == (9.0, 18.0)
assert doc.pages.get(3).size.as_tuple() == (13.0, 36.0)