From 6ab546626dbbefffa64d4f162e0599fa140d5c9d Mon Sep 17 00:00:00 2001
From: HuangHai <10402852@qq.com>
Date: Sat, 28 Jun 2025 10:36:03 +0800
Subject: [PATCH] 'commit'

---
 dsRag/Test/Test_MatchImage.py                 | 125 +++++++++++++++---
 dsRag/Util/WordImageUtil.py                   |  10 +-
 .../__pycache__/WordImageUtil.cpython-310.pyc | Bin 1713 -> 1766 bytes
 dsRag/static/Test/1.txt                       |   2 +-
 dsRag/static/Test/2.txt                       |   2 +-
 ...g => 81572c98254043d4a475ab7381979a67.png} | Bin
 ...g => b536401321764b20b3c000c120ab3c5b.png} | Bin
 7 files changed, 114 insertions(+), 25 deletions(-)
 rename dsRag/static/Test/{1.png => 81572c98254043d4a475ab7381979a67.png} (100%)
 rename dsRag/static/Test/{2.png => b536401321764b20b3c000c120ab3c5b.png} (100%)

diff --git a/dsRag/Test/Test_MatchImage.py b/dsRag/Test/Test_MatchImage.py
index 557d5e29..a05cbebe 100644
--- a/dsRag/Test/Test_MatchImage.py
+++ b/dsRag/Test/Test_MatchImage.py
@@ -1,9 +1,78 @@
-import os
-
+import re
 import docx
 
-from Util.WordImageUtil import extract_images_from_docx
-
+import os
+import shutil
+import uuid
+import zipfile
+
+from docx import Document
+from docx.oxml.ns import nsmap
+
+
+def extract_images_from_docx(docx_path, output_folder):
+    """
+    从docx提取图片并记录位置
+    :param docx_path: Word文档路径
+    :param output_folder: 图片输出文件夹
+    :return: 包含图片路径和位置的列表
+    """
+    # 创建一个List<String> 记录每个图片的名称和序号
+    image_data = []
+    # 创建临时解压目录
+    temp_dir = os.path.join(output_folder, "temp_docx")
+    os.makedirs(temp_dir, exist_ok=True)
+
+    # 解压docx文件
+    with zipfile.ZipFile(docx_path, 'r') as zip_ref:
+        zip_ref.extractall(temp_dir)
+
+    # 读取主文档关系
+    with open(os.path.join(temp_dir, 'word', '_rels', 'document.xml.rels'), 'r') as rels_file:
+        rels_content = rels_file.read()
+
+    # 加载主文档
+    doc = Document(docx_path)
+    img_counter = 1
+
+    # 遍历所有段落
+    for para_idx, paragraph in enumerate(doc.paragraphs):
+        for run_idx, run in enumerate(paragraph.runs):
+            # 检查运行中的图形
+            for element in run._element:
+                if element.tag.endswith('drawing'):
+                    # 提取图片关系ID
+                    blip = element.find('.//a:blip', namespaces=nsmap)
+                    if blip is not None:
+                        embed_id = blip.get('{%s}embed' % nsmap['r'])
+
+                        # 从关系文件中获取图片文件名
+                        rel_entry = f'<Relationship Id="{embed_id}"'
+                        if rel_entry in rels_content:
+                            start = rels_content.find(rel_entry)
+                            target_start = rels_content.find('Target="', start) + 8
+                            target_end = rels_content.find('"', target_start)
+                            image_path = rels_content[target_start:target_end]
+
+                            # 构建图片源路径
+                            src_path = os.path.join(temp_dir, 'word', image_path.replace('..', '').lstrip('/'))
+
+                            if os.path.exists(src_path):
+                                # 创建输出文件名
+                                ext = os.path.splitext(src_path)[1]
+                                # 名称为uuid
+                                fileName=uuid.uuid4().hex
+                                img_name = f"{fileName}{ext}"
+                                image_data.append(img_name)
+                                dest_path = os.path.join(output_folder, img_name)
+                                # 复制图片
+                                shutil.copy(src_path, dest_path)
+
+                                img_counter += 1
+
+    # 清理临时目录
+    shutil.rmtree(temp_dir)
+    return image_data
 
 def read_word_content(docx_path):
     res = ""
@@ -73,25 +142,43 @@ def save_to_txt(content, file_path, mode='w'):
         print(f"保存文件{file_path}时出错: {str(e)}")
         return False
 
+
 if __name__ == "__main__":
     word_document_path = "d:\\dsWork\\dsProject\\dsRag\\static\\Test\\带图的WORD文档_MATH_3.docx"
-    output_dir="d:\\dsWork\\dsProject\\dsRag\\static\\Test\\"
-    extract_images_from_docx(word_document_path, output_dir)
+    output_dir = "d:\\dsWork\\dsProject\\dsRag\\static\\Test\\"
+
+    # 提取图片
+    listImage = extract_images_from_docx(word_document_path, output_dir)
+    # 读取内容
     res = read_word_content(word_document_path)
+    # 分块
     chunks = split_into_blocks(res)
+    saved_count = 0
+
+    # 使用原来的正则表达式
+    pattern = re.compile(r'【图片\d+】')
+    # 初始化图片索引
+    img_idx = 0
+
     for x in chunks:
-        print("===段落开始：===")
         firstLine = x[1].split("\n")[0].strip()
         content = x[1][len(firstLine):].strip()
-        print("firstLine=" + firstLine)
-        print("content=" + content)
-        print("===段落结束：===\n")
-
-        saved_count=0
-        for chunk_num, chunk in chunks:
-            chunk = chunk.strip()  # 确保去除空白字符
-            output_file = os.path.join(output_dir, f"{chunk_num}.txt")
-            if save_to_txt(chunk, output_file, mode='w'):
-                saved_count += 1
-
-        print(f"处理完成，共保存{saved_count}个文件到目录: {output_dir}")
+
+        # 使用finditer查找所有匹配项
+        # 使用闭包函数替换所有匹配项
+        img_idx = [0]  # 使用列表实现可变状态
+        def replacer(match):
+            if img_idx[0] < len(listImage):
+                result = listImage[img_idx[0]]
+                img_idx[0] += 1
+                return result
+            return match.group()
+            
+        content = pattern.sub(replacer, content)
+        # 保存文本文件
+        output_file = os.path.join(output_dir, f"{x[0]}.txt")
+        full_content = f"{firstLine}\n{content}"
+        if save_to_txt(full_content, output_file, mode='w'):
+            saved_count += 1
+
+    print(f"处理完成，共保存{saved_count}个文件到目录: {output_dir}")
diff --git a/dsRag/Util/WordImageUtil.py b/dsRag/Util/WordImageUtil.py
index fc0e20e3..a887f482 100644
--- a/dsRag/Util/WordImageUtil.py
+++ b/dsRag/Util/WordImageUtil.py
@@ -14,6 +14,8 @@ def extract_images_from_docx(docx_path, output_folder):
     :param output_folder: 图片输出文件夹
     :return: 包含图片路径和位置的列表
     """
+    # 创建一个List<String> 记录每个图片的名称和序号
+    image_data = []
     # 创建临时解压目录
     temp_dir = os.path.join(output_folder, "temp_docx")
     os.makedirs(temp_dir, exist_ok=True)
@@ -30,8 +32,6 @@ def extract_images_from_docx(docx_path, output_folder):
     doc = Document(docx_path)
     img_counter = 1
 
-    idx = 0
-
     # 遍历所有段落
     for para_idx, paragraph in enumerate(doc.paragraphs):
         for run_idx, run in enumerate(paragraph.runs):
@@ -58,8 +58,9 @@ def extract_images_from_docx(docx_path, output_folder):
                                 # 创建输出文件名
                                 ext = os.path.splitext(src_path)[1]
                                 # 名称为uuid
-                                idx = idx + 1
-                                img_name = f"{idx}{ext}"
+                                fileName=uuid.uuid4().hex
+                                img_name = f"{fileName}{ext}"
+                                image_data.append(img_name)
                                 dest_path = os.path.join(output_folder, img_name)
                                 # 复制图片
                                 shutil.copy(src_path, dest_path)
@@ -68,3 +69,4 @@ def extract_images_from_docx(docx_path, output_folder):
 
     # 清理临时目录
     shutil.rmtree(temp_dir)
+    return image_data
diff --git a/dsRag/Util/__pycache__/WordImageUtil.cpython-310.pyc b/dsRag/Util/__pycache__/WordImageUtil.cpython-310.pyc
index 895377862d01e07dd15aff546e0bf3a991dac0e6..c91992519f8544fdc99cae9e46c21d500ad5ec8d 100644
GIT binary patch
delta 819
zcmZWn%Wl*#6tx{EllQ!5Izy=_s(@~aKm~Rzc&#dgRK#P{DvEGA4|N{RWMCH86lv6D
zsj4iiqGZXEFF^c8U{{IH0PHCes)%)u&pp@Y+Hvy5`RTYr$1#z3oM&F=v32ZzTy2WX
z67f*PIKv3F5hoGq;e-HYI3-*WnF7l3x9UCxnrNV;7^9bnsT5H(Q!pCnx6t_7JE6+)
zT11rl+d^R)%P~ErjLyiB%~ennJw;sm&;R6t-vBhjJjV>KM<(QY)(k&F``brbgd5xx
zlG(GRw0$5Hi`!uL9FcMEfYWo6JTp@qo(W!HR#as6A>l5#Or1#8O-jrWDs7+|N-@tf
z7xE{F<xdq}fC6Z5kyxcAf{SWVGFE_Mx=&cKj(BM<<e?-krwkgDpnO0Yh*$o-GgOi?
zE3tA)k_xM&I8}IcUadk^9&FBQk~2d%Yf(K}NKvoB>rg*IP&+}Vgf9R|ny?^m`@HjQ
zz#=8Q5nW0ap$d(ZU<5id?jf%4?d*AbFGS_?`3C~BiB_vICqHl&x0EG;@0BgQ6n|Hq
zmoszZhkiTsN1bs-!^vQr#SiF4gH9%0mZ=igGo=$uGtD23f&q&UY5jULGuZIW)EoIx
zr{xSM(P$ERJHuTTge|k%_uGNT{K${L()(6Rmpgb}HqBy3T^76FR9`Lcba#WNem`jG
z-G18}2qszv<7nRUGYZ$^PwHlKL;Q<WcZrK##m0-jo75DjQ;K6tyAqhh#wK>LO>{y&
zkTdejyh?F+SGZP<FfXR?hJ+OfWdRwL#}0|4wxnyrX@7TZFb);DL>j~Oi*P4C)K>lg
DO~lmM

delta 795
zcmY*W&ubGw6rP!#O=h>dlix{GMJ#w}m4XNHq6M)B6>k-*OC<zf`Xe+yI@#DBW{41a
z=%JRfcq)b6qbKnn5zoOQig@yGP~V0^afkQj`@Wg?&Fs8i{wH7deBUMb$tR7i_wI>&
z_n?k!2IC>d;4>v8ByAcqO@SuNFl3fyz`#0Ul!QFCH36QmW8YZXiZhy>e6Ngd>`_Y6
z@e(7<URW^5Ko&&8fsfg-XSn9ENQkl52-uu(GPWq`irZwc@?X89$(Wi9I3^3O5uvLy
z?U^xMA+HYCnQ@GZire<4$_SO2k@2+G_Ho0W@wMNUaSmn@s?HJT!3zuEA2B1f1bdjO
zGA@EdmDfp?B-G>}r*o$S@@JOG>%6w#5Zq<Qh;g-$A}HuW!k}0qra0w!UBve(X<HZd
zsC2;U#FQ`pTbJV!l%SljxD1tqCYGs8S1THsI-lZdiowCksutj&wb0^}=;|psW2ScD
zMs*Djnhk65tgh(U1lypq;M^r9ySuU5*nJ+>WL(!Q!4`nMN*1coA^S>$Xv4aeQP@xt
zL<iP7or`{0&q{&YZ>r{|YWBAV6Rw8cK@dIWPos~#dZivX(0je#=r_Zy!0!#i{xEE8
z^md@7qThTuKe4u9KM<)wBSq02dmhhOvtQJ<JDZJev(uV5&>DnOhYv(8MqljJxkq>f
zDI}GaN3Z^!<1S{wIpr+6pSeB~%%hTeOfdGA{bYaSyt;)I3y$(}Qr9s}_;9!ls(>)y
z=@k`Ap_js?-hO9isXI^>O4CZUn4-&c6X`_U>-1o_(^^g+WFRpKA}EgNTPFJtEM>-`

diff --git a/dsRag/static/Test/1.txt b/dsRag/static/Test/1.txt
index 672fbbbf..7667506d 100644
--- a/dsRag/static/Test/1.txt
+++ b/dsRag/static/Test/1.txt
@@ -2,4 +2,4 @@
 在教学过程中，引导学生构建和理解模型，不仅能提升他们分析和解决问题的能力，还能激发他们发现问题和提出问题的意识。例如，在认识路程模型时，教师可通过生活化情境让学生理解速度的概念及其单位表示。
 模型思想是《义务教育数学课程标准》中强调的核心素养之一，它帮助学生建立从现实世界抽象出数学问题的能力，并通过数学语言进行描述和解释。
 因此，在“综合与实践”类教学内容中，应加强模型的应用训练，以培养学生应用数学知识解决实际问题的能力。
-【图片1】
\ No newline at end of file
+81572c98254043d4a475ab7381979a67.png
\ No newline at end of file
diff --git a/dsRag/static/Test/2.txt b/dsRag/static/Test/2.txt
index a2f20a7c..fcab777c 100644
--- a/dsRag/static/Test/2.txt
+++ b/dsRag/static/Test/2.txt
@@ -1,2 +1,2 @@
 我随便写点什么
-【图片2】
\ No newline at end of file
+81572c98254043d4a475ab7381979a67.png
\ No newline at end of file
diff --git a/dsRag/static/Test/1.png b/dsRag/static/Test/81572c98254043d4a475ab7381979a67.png
similarity index 100%
rename from dsRag/static/Test/1.png
rename to dsRag/static/Test/81572c98254043d4a475ab7381979a67.png
diff --git a/dsRag/static/Test/2.png b/dsRag/static/Test/b536401321764b20b3c000c120ab3c5b.png
similarity index 100%
rename from dsRag/static/Test/2.png
rename to dsRag/static/Test/b536401321764b20b3c000c120ab3c5b.png