From ef3a9f10685d7c84cca386a67d7138ad02ae45dc Mon Sep 17 00:00:00 2001
From: "WIN-C27C5MSVTPO\\16961" <1696136552@qq.com>
Date: Tue, 16 Dec 2025 14:59:13 +0800
Subject: [PATCH] =?UTF-8?q?=E5=AF=BC=E5=85=A5?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 app/admin/controller/SkuController.php | 79 +++++++++++++-------------
 1 file changed, 41 insertions(+), 38 deletions(-)

diff --git a/app/admin/controller/SkuController.php b/app/admin/controller/SkuController.php
index 8f420fe..0544f19 100644
--- a/app/admin/controller/SkuController.php
+++ b/app/admin/controller/SkuController.php
@@ -264,10 +264,8 @@ class SkuController extends BaseController
             return '';
         }
 
-        // 1. 读取 docx
-        $phpWord = WordIOFactory::load($docxPath);
+        $phpWord = \PhpOffice\PhpWord\IOFactory::load($docxPath);
 
-        // 2. 导出 HTML
         $tmpDir = runtime_path('tmp/' . uniqid());
         if (!is_dir($tmpDir)) {
             mkdir($tmpDir, 0755, true);
@@ -275,48 +273,37 @@ class SkuController extends BaseController
 
         $tmpHtml = $tmpDir . '/index.html';
 
-        $writer = WordIOFactory::createWriter($phpWord, 'HTML');
+        $writer = \PhpOffice\PhpWord\IOFactory::createWriter($phpWord, 'HTML');
         $writer->save($tmpHtml);
 
         $html = file_get_contents($tmpHtml);
 
-        // 3. 只取 body 内容（关键）
-        if (preg_match('/<body[^>]*>(.*?)<\/body>/is', $html, $m)) {
-            $html = $m[1];
-        }
-
-        // 4. 清洗 Word 垃圾样式（关键）
-        $html = $this->cleanWordHtml($html);
-
-        // 5. 处理图片并上传 OSS
+        // 1️⃣ 上传图片到 OSS
         $uploadService = new UploadService();
-
         $html = preg_replace_callback(
             '/<img[^>]+src="([^"]+)"[^>]*>/i',
-            function ($matches) use ($tmpDir, $uploadService) {
-                $src = $matches[1];
+            function ($m) use ($tmpDir, $uploadService) {
+                $src = $m[1];
 
-                // Word 导出的是相对路径
                 if (!str_starts_with($src, 'http')) {
-                    $localImg = $tmpDir . '/' . ltrim($src, '/');
+                    $localImg = $tmpDir . '/' . ltrim($src, './');
                     if (is_file($localImg)) {
                         $ossUrl = $uploadService->uploadLocalFile([
-                            'path'  => $localImg,
+                            'path' => $localImg,
                             'scene' => 'desc'
                         ]);
-                        return preg_replace('/src="[^"]+"/', 'src="' . $ossUrl . '"', $matches[0]);
+                        return preg_replace('/src="[^"]+"/', 'src="'.$ossUrl['file_url'].'"', $m[0]);
                     }
                 }
-
-                return $matches[0];
+                return $m[0];
             },
             $html
         );
 
-        // 6. 清理临时目录
-        $this->deleteDir($tmpDir);
+        // 2️⃣ 关键：HTML 清洗（适配 wangEditor）
+        $html = $this->cleanWordHtmlForWangEditor($html);
 
-        return trim($html);
+        return $html;
     }
 
     function deleteDir(string $dir): void
@@ -325,28 +312,44 @@ class SkuController extends BaseController
         foreach (scandir($dir) as $file) {
             if ($file === '.' || $file === '..') continue;
             $path = $dir . '/' . $file;
-            is_dir($path) ? deleteDir($path) : @unlink($path);
+            is_dir($path) ? $this->deleteDir($path) : @unlink($path);
         }
         @rmdir($dir);
     }
 
-    function cleanWordHtml(string $html): string
+    function cleanWordHtmlForWangEditor(string $html): string
     {
-        // 移除 Office 命名空间、样式
-        $html = preg_replace('/\s*mso-[^:]+:[^;"]+;?/i', '', $html);
-        $html = preg_replace('/\s*class="[^"]*"/i', '', $html);
-        $html = preg_replace('/\s*style="[^"]*"/i', '', $html);
+        // 1. 移除 Word 专属标签
+        $html = preg_replace('/<o:p>.*?<\/o:p>/is', '', $html);
+        $html = preg_replace('/<\/?xml[^>]*>/i', '', $html);
+        $html = preg_replace('/<\/?v:[^>]*>/i', '', $html);
 
-        // 移除多余标签
-        $html = preg_replace('/<(\/?)(html|head|meta|link|span|font|xml|o:p)[^>]*>/i', '', $html);
+        // 2. 移除 class / style
+        $html = preg_replace('/\sclass="[^"]*"/i', '', $html);
+        $html = preg_replace('/\sstyle="[^"]*"/i', '', $html);
 
-        // 清理空段落
-        $html = preg_replace('/<p>\s*<\/p>/i', '', $html);
+        // 3. 清理 span（保留内容）
+        while (preg_match('/<span[^>]*>(.*?)<\/span>/is', $html)) {
+            $html = preg_replace('/<span[^>]*>(.*?)<\/span>/is', '$1', $html);
+        }
 
-        // 规范段落
-        $html = str_replace(['<p>', '</p>'], ["<p>", "</p>\n"], $html);
+        // 4. 转换 div → p
+        $html = preg_replace('/<div[^>]*>/i', '<p>', $html);
+        $html = preg_replace('/<\/div>/i', '</p>', $html);
 
-        return $html;
+        // 5. img 只保留 src
+        $html = preg_replace('/<img[^>]*src="([^"]+)"[^>]*>/i', '<img src="$1">', $html);
+
+        // 6. table 清洗
+        $html = preg_replace('/<table[^>]*>/i', '<table>', $html);
+        $html = preg_replace('/<td[^>]*>/i', '<td>', $html);
+        $html = preg_replace('/<th[^>]*>/i', '<th>', $html);
+
+        // 7. 去除多余空白
+        $html = preg_replace('/\s+/u', ' ', $html);
+        $html = str_replace(['<p></p>', '<p> </p>'], '', $html);
+
+        return trim($html);
     }