Skip to content

Commit fb2421d

Browse files
committed
fix: table rich cell hierarchy must have children for each element in parent
Signed-off-by: Christoph Auer <cau@zurich.ibm.com>
1 parent 215be28 commit fb2421d

1 file changed

Lines changed: 10 additions & 1 deletion

File tree

docling_cvat_tools/cvat_tools/cvat_to_docling.py

Lines changed: 10 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -1389,10 +1389,19 @@ def _process_table_data(self):
13891389
page_no = self.tabular_data[tind]["page_no"]
13901390
page_height = self.doc.pages[page_no].size.height
13911391

1392+
existing_table_children = list(table_item.children)
13921393
table_item.children = []
13931394
all_items = []
1395+
seen_item_refs: Set[str] = set()
1396+
for item_ref in existing_table_children:
1397+
if item_ref.cref not in seen_item_refs:
1398+
seen_item_refs.add(item_ref.cref)
1399+
all_items.append(item_ref)
13941400
for item, _ in self.doc.iterate_items(page_no=page_no):
1395-
all_items.append(item.get_ref())
1401+
item_ref = item.get_ref()
1402+
if item_ref.cref not in seen_item_refs:
1403+
seen_item_refs.add(item_ref.cref)
1404+
all_items.append(item_ref)
13961405

13971406
for c in table_cell_data:
13981407
# Get text to populate TableData

0 commit comments

Comments
 (0)