From 33d63c8efefe7ccb4fb7ac9cb71d4fb8a3af041f Mon Sep 17 00:00:00 2001
From: Henry-Shaw <34739790+Henry-Shaw@users.noreply.github.com>
Date: Wed, 16 Oct 2024 14:39:52 +0800
Subject: [PATCH] =?UTF-8?q?fix:=20=E4=BF=AE=E5=A4=8D=E7=9F=A5=E8=AF=86?=
 =?UTF-8?q?=E5=BA=93=E4=B8=8A=E4=BC=A0=E6=97=A7=E7=89=88=E6=9C=ACdocx?=
 =?UTF-8?q?=E6=96=87=E4=BB=B6=E5=90=8E=EF=BC=8C=E5=9B=BE=E7=89=87=E6=9C=AA?=
 =?UTF-8?q?=E6=AD=A3=E5=B8=B8=E8=AF=86=E5=88=AB=E5=AF=BC=E5=85=A5=E7=9A=84?=
 =?UTF-8?q?=E9=97=AE=E9=A2=98=20(#1382)?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 apps/common/handle/impl/doc_split_handle.py | 16 ++++++++++++++--
 1 file changed, 14 insertions(+), 2 deletions(-)

diff --git a/apps/common/handle/impl/doc_split_handle.py b/apps/common/handle/impl/doc_split_handle.py
index dd50d8887..ed65b5b73 100644
--- a/apps/common/handle/impl/doc_split_handle.py
+++ b/apps/common/handle/impl/doc_split_handle.py
@@ -15,6 +15,7 @@ from typing import List
 from docx import Document, ImagePart
 from docx.table import Table
 from docx.text.paragraph import Paragraph
+from docx.oxml import ns
 
 from common.handle.base_split_handle import BaseSplitHandle
 from common.util.split_model import SplitModel
@@ -27,9 +28,15 @@ default_pattern_list = [re.compile('(?<=^)# .*|(?<=\\n)# .*'),
                         re.compile("(?<=\\n)(?<!#)##### (?!#).*|(?<=^)(?<!#)##### (?!#).*"),
                         re.compile("(?<=\\n)(?<!#)###### (?!#).*|(?<=^)(?<!#)###### (?!#).*")]
 
+old_docx_nsmap = {'v': 'urn:schemas-microsoft-com:vml'}
+combine_nsmap = {**ns.nsmap, **old_docx_nsmap}
 
-def image_to_mode(image, doc: Document, images_list, get_image_id):
-    for img_id in image.xpath('.//a:blip/@r:embed'):  # 获取图片id
+def image_to_mode(image, doc: Document, images_list, get_image_id, is_new_docx=True):
+    if is_new_docx:
+        image_ids = image.xpath('.//a:blip/@r:embed')
+    else:
+        image_ids = image.xpath('.//v:imagedata/@r:id', namespaces=combine_nsmap)
+    for img_id in image_ids:  # 获取图片id
         part = doc.part.related_parts[img_id]  # 根据图片id获取对应的图片
         if isinstance(part, ImagePart):
             image_uuid = get_image_id(img_id)
@@ -42,10 +49,15 @@ def image_to_mode(image, doc: Document, images_list, get_image_id):
 def get_paragraph_element_txt(paragraph_element, doc: Document, images_list, get_image_id):
     try:
         images = paragraph_element.xpath(".//pic:pic")
+        old_docx_images = paragraph_element.xpath(".//w:pict")
         if len(images) > 0:
             return "".join(
                 [item for item in [image_to_mode(image, doc, images_list, get_image_id) for image in images] if
                  item is not None])
+        elif len(old_docx_images) > 0:
+            return "".join(
+                [item for item in [image_to_mode(image, doc, images_list, get_image_id, is_new_docx=False) for image in old_docx_images] if
+                 item is not None])
         elif paragraph_element.text is not None:
             return paragraph_element.text
         return ""