AI/ragflow
1
0
Fork 0
mirror of https://git.mirrors.martin98.com/https://github.com/infiniflow/ragflow.git synced 2025-08-14 16:25:54 +08:00
Code Issues Packages Projects Releases Wiki Activity

Fix: uncaptured image data with position information (#7683)

Browse Source 
### What problem does this PR solve?

Fixed uncaptured figure data with position information. #7466, #7681

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)

Co-authored-by: Kevin Hu <kevinhu.sh@gmail.com>
This commit is contained in:
Yongteng Lei 2025-05-19 19:33:28 +08:00 committed by GitHub
parent 0ebf05440e
commit b908c33464
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 14 additions and 13 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

27
deepdoc/parser/figure_parser.py
View File

@ -22,13 +22,19 @@ from rag.prompts import vision_llm_figure_describe_prompt
def vision_figure_parser_figure_data_wraper(figures_data_without_positions): def vision_figure_parser_figure_data_wraper(figures_data_without_positions):
return [( return [
(figure_data[1], [figure_data[0]]), (
[(0, 0, 0, 0, 0)] (figure_data[1], [figure_data[0]]),
) for figure_data in figures_data_without_positions if isinstance(figure_data[1], Image.Image)] [(0, 0, 0, 0, 0)],
)
for figure_data in figures_data_without_positions
if isinstance(figure_data[1], Image.Image)
]
shared_executor = ThreadPoolExecutor(max_workers=10) shared_executor = ThreadPoolExecutor(max_workers=10)
class VisionFigureParser: class VisionFigureParser:
def __init__(self, vision_model, figures_data, *args, **kwargs): def __init__(self, vision_model, figures_data, *args, **kwargs):
self.vision_model = vision_model self.vision_model = vision_model
@ -43,21 +49,16 @@ class VisionFigureParser:
for item in figures_data: for item in figures_data:
# position # position
if len(item) == 2 and isinstance(item[1], list) and len(item[1]) == 1 and isinstance(item[1][0], tuple) and len(item[1][0]) == 5: if len(item) == 2 and isinstance(item[0], tuple) and len(item[0]) == 2 and isinstance(item[1], list) and isinstance(item[1][0], tuple) and len(item[1][0]) == 5:
img_desc = item[0] img_desc = item[0]
assert len(img_desc) == 2 and isinstance(img_desc[0], Image.Image) and isinstance(img_desc[1], list), "Should be (figure, [description])" assert len(img_desc) == 2 and isinstance(img_desc[0], Image.Image) and isinstance(img_desc[1], list), "Should be (figure, [description])"
self.figures.append(img_desc[0]) self.figures.append(img_desc[0])
self.descriptions.append(img_desc[1]) self.descriptions.append(img_desc[1])
self.positions.append(item[1]) self.positions.append(item[1])
else: else:
assert len(item) == 2 and isinstance(item, tuple) and isinstance(item[1], list), f"get {len(item)=}, {item=}" assert len(item) == 2 and isinstance(item[0], Image.Image) and isinstance(item[1], list), f"Unexpected form of figure data: get {len(item)=}, {item=}"
if isinstance(item[0], tuple) and len(item[0]) == 2 and isinstance(item[0][0], Image.Image) and isinstance(item[0][1], list): self.figures.append(item[0])
self.figures.append(item[0][0]) self.descriptions.append(item[1])
self.descriptions.append(item[0][1])
else:
self.figures.append(item[0])
self.descriptions.append(item[1])
def _assemble(self): def _assemble(self):
self.assembled = [] self.assembled = []