From bd3f6e4a9bbae81ca972e81a0568cee16a3cb46e Mon Sep 17 00:00:00 2001
From: shaohuzhang1 <shaohu.zhang@fit2cloud.com>
Date: Wed, 10 Apr 2024 10:38:17 +0800
Subject: [PATCH] =?UTF-8?q?fix:=20word=E5=88=86=E6=AE=B5=E6=94=AF=E6=8C=81?=
 =?UTF-8?q?=E8=A1=A8=E6=A0=BC=E6=95=B0=E6=8D=AE?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 apps/common/handle/impl/doc_split_handle.py | 28 +++++++++++++++++++--
 1 file changed, 26 insertions(+), 2 deletions(-)

diff --git a/apps/common/handle/impl/doc_split_handle.py b/apps/common/handle/impl/doc_split_handle.py
index d3002b37a..d1f538936 100644
--- a/apps/common/handle/impl/doc_split_handle.py
+++ b/apps/common/handle/impl/doc_split_handle.py
@@ -11,6 +11,8 @@ import re
 from typing import List
 
 from docx import Document
+from docx.table import Table
+from docx.text.paragraph import Paragraph
 
 from common.handle.base_split_handle import BaseSplitHandle
 from common.util.split_model import SplitModel
@@ -34,9 +36,31 @@ class DocSplitHandle(BaseSplitHandle):
             return paragraph.text
         return paragraph.text
 
+    @staticmethod
+    def table_to_md(table):
+        rows = table.rows
+        # 创建 Markdown 格式的表格
+        md_table = '| ' + ' | '.join([cell.text.replace("\n", '</br>') for cell in rows[0].cells]) + ' |\n'
+        md_table += '| ' + ' | '.join(['---' for i in range(len(rows[0].cells))]) + ' |\n'
+        for row in rows[1:]:
+            md_table += '| ' + ' | '.join([cell.text.replace("\n", '</br>') for cell in row.cells]) + ' |\n'
+        return md_table
+
     def to_md(self, doc):
-        ps = doc.paragraphs
-        return "\n".join([self.paragraph_to_md(para) for para in ps])
+        elements = []
+        for element in doc.element.body:
+            if element.tag.endswith('tbl'):
+                # 处理表格
+                table = Table(element, doc)
+                elements.append(table)
+            elif element.tag.endswith('p'):
+                # 处理段落
+                paragraph = Paragraph(element, doc)
+                elements.append(paragraph)
+
+        return "\n".join(
+            [self.paragraph_to_md(element) if isinstance(element, Paragraph) else self.table_to_md(element) for element
+             in elements])
 
     def handle(self, file, pattern_list: List, with_filter: bool, limit: int, get_buffer):
         try: