docling-project / docling-project/docling
[BEE] Table model missing the last part of the table in the Japanese document
Open
bug
- Dominant language
- Python
- Stars
- 66.4k
- Forks
- 4.8k
- Avg merge
- 2d 21h
- Merged PRs (30d)
- 84
Description
### Bug
The bottom row of the table is missing from the output.
Observation:
These are selectable texts in the table
The token appears in the token list but not considered in table structure.
As per bounding box , token is well within the page
{
"id": "TOKEN_000102",
"confidence": 1.0,
"bbox": { "x": 169, "y": 606, "width": 110, "height": 10, "page_number": 1 },
"text": "ESGワーキング・グループ"
}
### Steps to reproduce
Run IBM_2021_ESG_Report 10.pdf through the latest docling.
The document is kept here -> https://ibm.ent.box.com/folder/380274980502
### Docling version
Latest
### Python version
3.12
Contributor guide
Assessment
This issue has not been assessed yet.