表格问题修复

table/table_cleaner.py

@@ -772,6 +772,7 @@ class DocCleaner:
         # 分析表格结构（查找合并单元格）
         merged_cells = {}
         merged_v_cells = set()  # 记录被垂直合并的单元格
+        cell_map = {}  # 添加cell_map的定义
         
         # 检测合并单元格
         for i in range(rows):
@@ -799,49 +800,106 @@ class DocCleaner:
                 except Exception as e:
                     print(f"警告：处理合并单元格时出错 [{i},{j}]: {str(e)}")
         
-        # 第二遍：复制内容并执行合并
-        for i in range(rows):
-            for j in range(cols):
+        # 构建HTML表格
+        html = f'<table id="{table_id}" class="docx-table">\n'
+        html += '<thead>\n'
+        
+        # 添加表头行
+        header_rows = min(1, rows)  # 假设第一行是表头
+        for i in range(header_rows):
+            html += '  <tr>\n'
+            j = 0
+            while j < cols:
                 try:
-                    src_cell = table.cell(i, j)
-                    dst_cell = new_table.cell(i, j)
+                    cell = table.cell(i, j)
+                    text = cell.text.strip()
+                    
+                    # 检查是否是合并单元格
+                    rowspan = 1
+                    colspan = 1
                     
-                    # 检查是否需要合并
                     if (i, j) in cell_map:
                         merge_type, span = cell_map[(i, j)]
                         if merge_type == 'vmerge':
-                            # 垂直合并
+                            rowspan = span
+                        elif merge_type == 'hmerge':
+                            colspan = span
+                    
+                    # 添加表头单元格
+                    attrs = []
+                    if rowspan > 1:
+                        attrs.append(f'rowspan="{rowspan}"')
+                    if colspan > 1:
+                        attrs.append(f'colspan="{colspan}"')
+                    
+                    attrs_str = ' '.join(attrs)
+                    if attrs_str:
+                        attrs_str = ' ' + attrs_str
+                    
+                    html += f'    <th{attrs_str}>{text}</th>\n'
+                    
+                    # 如果是水平合并，跳过合并的列
+                    j += colspan
+                except Exception as e:
+                    print(f"警告：处理表头单元格时出错 [{i},{j}]: {str(e)}")
+                    html += f'    <th>错误: {str(e)}</th>\n'
+                    j += 1
+            html += '  </tr>\n'
+        
+        html += '</thead>\n<tbody>\n'
+        
+        # 添加数据行
+        for i in range(header_rows, rows):
+            html += '  <tr>\n'
+            j = 0
+            while j < cols:
+                try:
+                    # 跳过已经被垂直合并的单元格
+                    if (i, j) in merged_v_cells:
+                        j += 1
+                        continue
+                    
+                    cell = table.cell(i, j)
+                    text = cell.text.strip()
+                    
+                    # 检查是否是合并单元格
+                    rowspan = 1
+                    colspan = 1
+                    
+                    if (i, j) in cell_map:
+                        merge_type, span = cell_map[(i, j)]
+                        if merge_type == 'vmerge':
+                            rowspan = span
+                            # 标记被垂直合并的单元格
                             for k in range(1, span):
                                 if i + k < rows:
-                                    dst_cell.merge(new_table.cell(i + k, j))
+                                    merged_v_cells.add((i + k, j))
                         elif merge_type == 'hmerge':
-                            # 水平合并
-                            for k in range(1, span):
-                                if j + k < cols:
-                                    dst_cell.merge(new_table.cell(i, j + k))
+                            colspan = span
                     
-                    # 复制单元格属性
-                    if src_cell._element.tcPr is not None:
-                        dst_cell._element.tcPr = deepcopy(src_cell._element.tcPr)
+                    # 添加数据单元格
+                    attrs = []
+                    if rowspan > 1:
+                        attrs.append(f'rowspan="{rowspan}"')
+                    if colspan > 1:
+                        attrs.append(f'colspan="{colspan}"')
                     
-                    # 复制单元格内容
-                    dst_cell.text = ""  # 清除默认内容
-                    for src_paragraph in src_cell.paragraphs:
-                        dst_paragraph = dst_cell.add_paragraph()
-                        # 复制段落属性
-                        if src_paragraph._element.pPr is not None:
-                            dst_paragraph._element.pPr = deepcopy(src_paragraph._element.pPr)
-                        
-                        # 复制文本和格式
-                        for src_run in src_paragraph.runs:
-                            dst_run = dst_paragraph.add_run(src_run.text)
-                            # 复制运行属性
-                            if src_run._element.rPr is not None:
-                                dst_run._element.rPr = deepcopy(src_run._element.rPr)
-                
+                    attrs_str = ' '.join(attrs)
+                    if attrs_str:
+                        attrs_str = ' ' + attrs_str
+                    
+                    html += f'    <td{attrs_str}>{text}</td>\n'
+                    
+                    # 如果是水平合并，跳过合并的列
+                    j += colspan
                 except Exception as e:
-                    print(f"警告：复制单元格时出错 [{i},{j}]: {str(e)}")
-                    continue
+                    print(f"警告：处理数据单元格时出错 [{i},{j}]: {str(e)}")
+                    html += f'    <td>错误: {str(e)}</td>\n'
+                    j += 1
+            html += '  </tr>\n'
+        
+        html += '</tbody>\n</table>'
+        return html
 
     def _get_vmerge_value(self, cell_element) -> str:
         """