docling-project / docling-project/docling
Hierarchical Chunker TripletTableSerializer joins table rows into one massive block
Open
bug
- Dominant language
- Python
- Stars
- 66.4k
- Forks
- 4.8k
- Avg merge
- 2d 21h
- Merged PRs (30d)
- 84
Description
### Bug
Given a table
| column_a | column_b | column_c |
|--|--|--|
| a | b | c |
| d | e | f |
The TripletTableSerializer turns it into
```
a, column_b = b. column_c = c. d, column_b = e. column_c = f.
```
Ideally the columns would be separated by commas and the rows would be on different lines or at least using a different separator
Contributor guide
Assessment
This issue has not been assessed yet.