From 56def59c2b8889122275b1cf65384e726d5b8c5c Mon Sep 17 00:00:00 2001
From: Stephen Hu <812791840@qq.com>
Date: Mon, 27 Oct 2025 13:23:16 +0800
Subject: [PATCH] Fix:Error retrieving DOCX image
 (docx.image.exceptions.UnrecognizedImageError)  (#10794)

### What problem does this PR solve?

https://github.com/infiniflow/ragflow/issues/10776

### Type of change

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

---------

Co-authored-by: Kevin Hu <kevinhu.sh@gmail.com>
---
 rag/app/manual.py | 21 +++++++++++++++------
 1 file changed, 15 insertions(+), 6 deletions(-)

diff --git a/rag/app/manual.py b/rag/app/manual.py
index 32a38ed39..a433a10e2 100644
--- a/rag/app/manual.py
+++ b/rag/app/manual.py
@@ -80,12 +80,21 @@ class Docx(DocxParser):
         img = paragraph._element.xpath('.//pic:pic')
         if not img:
             return None
-        img = img[0]
-        embed = img.xpath('.//a:blip/@r:embed')[0]
-        related_part = document.part.related_parts[embed]
-        image = related_part.image
-        image = Image.open(BytesIO(image.blob))
-        return image
+        try:
+            img = img[0]
+            embed = img.xpath('.//a:blip/@r:embed')[0]
+            related_part = document.part.related_parts[embed]
+            image = related_part.image
+            if image is not None:
+                image = Image.open(BytesIO(image.blob))
+                return image
+            elif related_part.blob is not None:
+                image = Image.open(BytesIO(related_part.blob))
+                return image
+            else:
+                return None
+        except Exception:
+            return None
 
     def concat_img(self, img1, img2):
         if img1 and not img2: