docling-project / docling-project/docling

[Bee] Rotated table is not correctly extracted

Open
#2,343 2 comments 0 reactions 0 assignees View on GitHub
enhancement
Dominant language
Python
Stars
66.4k
Forks
4.8k
Avg merge
3d 4h
Merged PRs (30d)
95

Description

The page contains a rotated table. The table is identified, but not rotated back, the columns are recognized as row.
Page rotation is 0.

**Steps to reproduce:**
use the docling pipeline to process this PDF file and export the table

**Expected behavior:**
The table should be rotated back, the recognized row and column should follow the correct reading order.

**Test file:**
[2023-Annual-Report-Page48.pdf](https://github.com/user-attachments/files/22222372/2023-Annual-Report-Page48.pdf)

**HTML output:**
[2023-Annual-Report-Page48-table-1.html](https://github.com/user-attachments/files/22222407/2023-Annual-Report-Page48-table-1.html)

**Docling version:**
docling==2.51.0
docling-core==2.47.0
docling-ibm-models==3.9.1
docling-parse==4.4.0

Contributor guide

Open the contributing guide

Research direction

Start by reproducing the issue with the docling pipeline using 2023-Annual-Report-Page48.pdf and compare the exported table with the provided HTML output. Trace the table extraction and rotation handling, then verify that the recognized rows and columns follow the correct reading order for this page.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
data
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.