mirrors/docling
1
0
Fork 0
mirror of https://github.com/DS4SD/docling.git synced 2025-07-24 19:14:23 +00:00
Code Issues Packages Projects Releases Wiki Activity

fix lint

Browse Source
This commit is contained in:
Qiefan Jiang 2025-07-03 10:27:52 +08:00
parent 37a0bf8e4c
commit b6b5b090a9
1 changed files with 18 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

27
docling/backend/msexcel_backend.py
View File

@ -337,13 +337,16 @@ class MsExcelDocumentBackend(DeclarativeDocumentBackend, PaginatedDocumentBacken
# Collect the data within the bounds # Collect the data within the bounds
data = [] data = []
visited_cells: set[tuple[int, int]] = set() visited_cells: set[tuple[int, int]] = set()
for ri, row in enumerate(sheet.iter_rows( for ri, row in enumerate(
sheet.iter_rows(
min_row=start_row + 1, # start_row is 0-based but iter_rows is 1-based min_row=start_row + 1, # start_row is 0-based but iter_rows is 1-based
max_row=max_row + 1, max_row=max_row + 1,
min_col=start_col + 1, min_col=start_col + 1,
max_col=max_col + 1, max_col=max_col + 1,
values_only=False values_only=False,
), start_row): ),
start_row,
):
for rj, cell in enumerate(row, start_col): for rj, cell in enumerate(row, start_col):
# Check if the cell belongs to a merged range # Check if the cell belongs to a merged range
row_span = 1 row_span = 1
@ -401,13 +404,16 @@ class MsExcelDocumentBackend(DeclarativeDocumentBackend, PaginatedDocumentBacken
""" """
max_row: int = start_row max_row: int = start_row
for ri, (cell,) in enumerate(sheet.iter_rows( for ri, (cell,) in enumerate(
sheet.iter_rows(
min_row=start_row + 2, min_row=start_row + 2,
max_row=sheet.max_row, max_row=sheet.max_row,
min_col=start_col + 1, min_col=start_col + 1,
max_col=start_col + 1, max_col=start_col + 1,
values_only=False values_only=False,
), start_row + 1): ),
start_row + 1,
):
# Check if the cell is part of a merged range # Check if the cell is part of a merged range
merged_range = next( merged_range = next(
(mr for mr in sheet.merged_cells.ranges if cell.coordinate in mr), (mr for mr in sheet.merged_cells.ranges if cell.coordinate in mr),
@ -440,13 +446,16 @@ class MsExcelDocumentBackend(DeclarativeDocumentBackend, PaginatedDocumentBacken
""" """
max_col: int = start_col max_col: int = start_col
for rj, (cell,) in enumerate(sheet.iter_cols( for rj, (cell,) in enumerate(
sheet.iter_cols(
min_row=start_row + 1, min_row=start_row + 1,
max_row=start_row + 1, max_row=start_row + 1,
min_col=start_col + 2, min_col=start_col + 2,
max_col=sheet.max_column, max_col=sheet.max_column,
values_only=False values_only=False,
), start_col + 1): ),
start_col + 1,
):
# Check if the cell is part of a merged range # Check if the cell is part of a merged range
merged_range = next( merged_range = next(
(mr for mr in sheet.merged_cells.ranges if cell.coordinate in mr), (mr for mr in sheet.merged_cells.ranges if cell.coordinate in mr),