From 5aae73c230d453d033bdfab01c437129b27f3a83 Mon Sep 17 00:00:00 2001
From: dylan <xuhao296@qq.com>
Date: Mon, 14 Apr 2025 10:10:20 +0800
Subject: [PATCH] Make error messages during PPT processing clearer. (#6980)

### What problem does this PR solve?

Sometimes a slide may trigger a Proxy error (ArgumentException:
Parameter is not valid) due to issues in the original file, and this
error message can be confusing for users.

### Type of change

- [ ] Bug Fix (non-breaking change which fixes an issue)
- [ ] New Feature (non-breaking change which adds functionality)
- [ ] Documentation Update
- [ ] Refactoring
- [ ] Performance Improvement
- [x] Other (please describe):
---
 rag/app/presentation.py | 13 ++++++++-----
 1 file changed, 8 insertions(+), 5 deletions(-)

diff --git a/rag/app/presentation.py b/rag/app/presentation.py
index fcfabd39..741470d5 100644
--- a/rag/app/presentation.py
+++ b/rag/app/presentation.py
@@ -36,11 +36,14 @@ class Ppt(PptParser):
         imgs = []
         with slides.Presentation(BytesIO(fnm)) as presentation:
             for i, slide in enumerate(presentation.slides[from_page: to_page]):
-                buffered = BytesIO()
-                slide.get_thumbnail(
-                    0.5, 0.5).save(
-                    buffered, drawing.imaging.ImageFormat.jpeg)
-                imgs.append(Image.open(buffered))
+                try:
+                    buffered = BytesIO()
+                    slide.get_thumbnail(
+                        0.5, 0.5).save(
+                        buffered, drawing.imaging.ImageFormat.jpeg)
+                    imgs.append(Image.open(buffered))
+                except RuntimeError as e:
+                    raise RuntimeError(f'ppt parse error at page {i+1}, original error: {str(e)}') from e
         assert len(imgs) == len(
             txts), "Slides text and image do not match: {} vs. {}".format(len(imgs), len(txts))
         callback(0.9, "Image extraction finished")