feat(export):添加页面设置配置和文档网格支持

添加页边距配置（上、下、左、右、装订线），支持 twips 到 EMU 单位转换
添加纸张尺寸设置（页面宽度、页面高度）支持
添加页面方向控制（纵向/横向）
添加页眉页脚距离配置
添加首页页眉页脚不同设置支持
添加文档网格设置（网格类型、每行字符数、每页行数）
实现基于 XML 的文档网格配置功能

app/services/export_service.py

@@ -13,7 +13,8 @@ from docx.enum.table import WD_TABLE_ALIGNMENT, WD_CELL_VERTICAL_ALIGNMENT
 from docx.enum.text import WD_ALIGN_PARAGRAPH
 from docx.oxml import OxmlElement
 from docx.oxml.ns import qn
-from docx.shared import Pt, RGBColor
+from docx.shared import Pt, RGBColor, Inches
+from docx.enum.section import WD_ORIENT
 from lxml import etree
 
 from app.config import settings
@@ -272,6 +273,157 @@ def _apply_run_style(run, style: dict):
 
 
 # ------------------------------------------------------------------ #
+# 页面设置应用
+# ------------------------------------------------------------------ #
+
+def twips_to_emu(twips: int) -> int:
+    """twips 转 EMU (1 twips = 635 EMU)"""
+    if twips is None:
+        return None
+    return int(twips * 635)
+
+
+def apply_page_setup(doc: Document, style_data: dict) -> None:
+    """应用页面设置到文档
+    
+    从 style_data 中读取 page_setup，应用到文档的第一个 section
+    
+    Args:
+        doc: python-docx Document 对象
+        style_data: 样式数据（包含 page_setup）
+    """
+    page_setup = style_data.get("page_setup")
+    if not page_setup:
+        return
+    
+    sections = page_setup.get("sections", [])
+    if not sections:
+        return
+    
+    # 应用第一节的设置
+    section_data = sections[0]
+    
+    # 检查文档是否有 section
+    if not doc.sections:
+        return
+    
+    section = doc.sections[0]
+    
+    try:
+        # 页边距（twips → EMU）
+        top_margin = section_data.get("top_margin")
+        if top_margin is not None:
+            section.top_margin = twips_to_emu(top_margin)
+        
+        bottom_margin = section_data.get("bottom_margin")
+        if bottom_margin is not None:
+            section.bottom_margin = twips_to_emu(bottom_margin)
+        
+        left_margin = section_data.get("left_margin")
+        if left_margin is not None:
+            section.left_margin = twips_to_emu(left_margin)
+        
+        right_margin = section_data.get("right_margin")
+        if right_margin is not None:
+            section.right_margin = twips_to_emu(right_margin)
+        
+        gutter = section_data.get("gutter")
+        if gutter is not None and gutter > 0:
+            section.gutter = twips_to_emu(gutter)
+        
+        # 纸张尺寸（twips → EMU）
+        page_width = section_data.get("page_width")
+        if page_width is not None:
+            section.page_width = twips_to_emu(page_width)
+        
+        page_height = section_data.get("page_height")
+        if page_height is not None:
+            section.page_height = twips_to_emu(page_height)
+        
+        # 方向
+        orientation = section_data.get("orientation")
+        if orientation == "landscape":
+            section.orientation = WD_ORIENT.LANDSCAPE
+        elif orientation == "portrait":
+            section.orientation = WD_ORIENT.PORTRAIT
+        
+        # 页眉页脚距离（twips → EMU）
+        header_distance = section_data.get("header_distance")
+        if header_distance is not None:
+            section.header_distance = twips_to_emu(header_distance)
+        
+        footer_distance = section_data.get("footer_distance")
+        if footer_distance is not None:
+            section.footer_distance = twips_to_emu(footer_distance)
+        
+        # 首页页眉页脚不同
+        different_first_page = section_data.get("different_first_page")
+        if different_first_page is not None:
+            section.different_first_page_header_footer = different_first_page
+        
+        # 文档网格（需要通过 XML 操作）
+        grid_type = section_data.get("grid_type")
+        chars_per_line = section_data.get("chars_per_line")
+        lines_per_page = section_data.get("lines_per_page")
+        
+        if grid_type or chars_per_line or lines_per_page:
+            _apply_document_grid(section, grid_type, chars_per_line, lines_per_page)
+        
+    except Exception as e:
+        # 如果应用页面设置失败，不影响文档生成，只是可能使用默认设置
+        print(f"警告: 应用页面设置失败: {e}")
+        pass
+
+
+def _apply_document_grid(section, grid_type: str = None, chars_per_line: int = None, lines_per_page: int = None) -> None:
+    """应用文档网格设置（通过 XML 操作）
+    
+    Args:
+        section: python-docx Section 对象
+        grid_type: 网格类型（default/lines/linesAndChars/snapToChars）
+        chars_per_line: 每行字符数
+        lines_per_page: 每页行数
+    """
+    try:
+        # 获取 section 的 XML 元素
+        sectPr = None
+        if hasattr(section, '_sectPr'):
+            sectPr = section._sectPr
+        elif hasattr(section, '_element'):
+            sectPr = section._element
+        
+        if sectPr is None:
+            return
+        
+        # 查找或创建 docGrid 元素
+        docGrid = sectPr.find(qn('w:docGrid'))
+        
+        if docGrid is None:
+            # 如果不存在，创建新的 docGrid 元素
+            docGrid = OxmlElement('w:docGrid')
+            # 插入到合适的位置（在 sectPr 的子元素中）
+            sectPr.append(docGrid)
+        
+        # 设置网格类型
+        if grid_type:
+            docGrid.set(qn('w:type'), grid_type)
+        
+        # 设置每页行数（linePitch）
+        # 注意：Word XML 中 linePitch 表示行间距，用于控制每页行数
+        if lines_per_page is not None and lines_per_page > 0:
+            docGrid.set(qn('w:linePitch'), str(lines_per_page))
+        
+        # 设置每行字符数（charSpace）
+        # 注意：Word XML 中 charSpace 表示字符间距，用于控制每行字符数
+        if chars_per_line is not None and chars_per_line > 0:
+            docGrid.set(qn('w:charSpace'), str(chars_per_line))
+        
+    except Exception as e:
+        print(f"警告: 应用文档网格失败: {e}")
+        pass
+
+
+# ------------------------------------------------------------------ #
 # Blocks → Word 转换
 # ------------------------------------------------------------------ #
 
@@ -287,8 +439,13 @@ def blocks_to_docx_bytes(blocks: list[dict], style_map: dict, style_data: dict)
         Word 文档字节流
     """
     doc = Document()
+    
+    # 注入样式
     inject_styles_from_json(doc, style_data)
     
+    # 应用页面设置（在注入样式之后）
+    apply_page_setup(doc, style_data)
+    
     # 验证并修正 Normal 样式的段后间距
     # 这是为了确保样式正确应用，避免 python-docx 的默认值覆盖
     _fix_normal_style_spacing(doc, style_data)

tmp/default.json

@@ -30,7 +30,7 @@
         "first_line_indent_pt": 44.0,
         "space_before_pt": null,
         "space_after_pt": null,
-        "line_spacing": 1.5,
+        "line_spacing": 1.0,
         "keep_together": null,
         "keep_with_next": null,
         "page_break_before": null
@@ -72,7 +72,7 @@
               "{http://schemas.openxmlformats.org/wordprocessingml/2006/main}spacing": {
                 "@tag": "{http://schemas.openxmlformats.org/wordprocessingml/2006/main}spacing",
                 "@attrib": {
-                  "{http://schemas.openxmlformats.org/wordprocessingml/2006/main}line": "360",
+                  "{http://schemas.openxmlformats.org/wordprocessingml/2006/main}line": "240",
                   "{http://schemas.openxmlformats.org/wordprocessingml/2006/main}lineRule": "auto"
                 }
               },
@@ -4799,5 +4799,26 @@
         }
       ]
     }
+  },
+  "page_setup": {
+    "sections": [
+      {
+        "top_margin": 1417,
+        "bottom_margin": 1134,
+        "left_margin": 1417,
+        "right_margin": 1134,
+        "gutter": 0,
+        "page_width": 11906,
+        "page_height": 16838,
+        "orientation": "portrait",
+        "header_distance": 851,
+        "footer_distance": 992,
+        "different_first_page": false,
+        "grid_type": "lines",
+        "chars_per_line": 0,
+        "lines_per_page": 312,
+        "paper_size_inferred": "A4"
+      }
+    ]
   }
 }

tmp/styles.py

@@ -1,6 +1,7 @@
 """
 提取 Word 文档中所有样式的完整 XML 定义（不遗漏任何属性）。
 同时提取编号格式定义（numbering.xml）以支持标题编号。
+同时提取页面设置（页边距、纸张大小、方向、页眉页脚等）。
 目标文件: default.docx
 """
 
@@ -22,6 +23,13 @@ def emu_to_pt(emu) -> float | None:
     return round(int(emu) / 12700, 2)
 
 
+def emu_to_twips(emu) -> int | None:
+    """EMU 转 twips (1 twips = 635 EMU)"""
+    if emu is None:
+        return None
+    return int(emu / 635)
+
+
 def extract_font(style) -> dict | None:
     """通过 API 提取字体信息（作为便捷摘要）"""
     try:
@@ -184,6 +192,152 @@ def extract_numbering(doc_path: Path) -> dict | None:
         return None
 
 
+def infer_paper_size(width_twips: int, height_twips: int) -> str:
+    """根据页面尺寸推断纸张类型"""
+    # 常见纸张尺寸（twips）
+    # A4 (纵向): 210mm x 297mm = 11906 x 16838 twips
+    # A4 (横向): 297mm x 210mm = 16838 x 11906 twips
+    # Letter (纵向): 8.5" x 11" = 12240 x 15840 twips
+    # A3 (纵向): 297mm x 420mm = 16838 x 23811 twips
+    
+    # 允许 ±100 twips 的误差
+    tolerance = 100
+    
+    # A4 纵向
+    if abs(width_twips - 11906) < tolerance and abs(height_twips - 16838) < tolerance:
+        return "A4"
+    # A4 横向
+    elif abs(width_twips - 16838) < tolerance and abs(height_twips - 11906) < tolerance:
+        return "A4 (Landscape)"
+    # Letter 纵向
+    elif abs(width_twips - 12240) < tolerance and abs(height_twips - 15840) < tolerance:
+        return "Letter"
+    # A3 纵向
+    elif abs(width_twips - 16838) < tolerance and abs(height_twips - 23811) < tolerance:
+        return "A3"
+    else:
+        return "Custom"
+
+
+def extract_page_setup(doc_path: Path) -> dict:
+    """
+    提取页面设置信息（页边距、纸张大小、方向、页眉页脚等）
+    只提取第一个节(section)的设置（大多数文档只有一个节）
+    """
+    try:
+        doc = Document(str(doc_path))
+        
+        # 检查是否有 section
+        if not doc.sections:
+            return {"sections": []}
+        
+        # 只提取第一个 section
+        section = doc.sections[0]
+        sections_data = []
+        # 提取页边距（转换为 twips）
+        top_margin = emu_to_twips(section.top_margin)
+        bottom_margin = emu_to_twips(section.bottom_margin)
+        left_margin = emu_to_twips(section.left_margin)
+        right_margin = emu_to_twips(section.right_margin)
+        gutter = emu_to_twips(section.gutter)
+        
+        # 提取纸张尺寸（转换为 twips）
+        page_width = emu_to_twips(section.page_width)
+        page_height = emu_to_twips(section.page_height)
+        
+        # 提取方向
+        # orientation: 0 = PORTRAIT, 1 = LANDSCAPE
+        orientation = "portrait" if section.orientation == 0 else "landscape"
+        
+        # 提取页眉页脚距离（转换为 twips）
+        header_distance = emu_to_twips(section.header_distance)
+        footer_distance = emu_to_twips(section.footer_distance)
+        
+        # 提取首页不同设置
+        different_first_page = section.different_first_page_header_footer
+        
+        # 从 XML 提取文档网格设置
+        # 注意：需要通过 section._sectPr 访问 XML 元素
+        grid_type = None
+        chars_per_line = None
+        lines_per_page = None
+        
+        try:
+            # 尝试获取 section 的 XML 元素
+            if hasattr(section, '_sectPr'):
+                sectPr = section._sectPr
+            elif hasattr(section, 'element'):
+                sectPr = section.element
+            else:
+                sectPr = None
+            
+            if sectPr is not None:
+                docGrid = sectPr.find(qn('w:docGrid'))
+                
+                if docGrid is not None:
+                    # 网格类型: default, lines, linesAndChars, snapToChars
+                    grid_type = docGrid.get(qn('w:type'))
+                    
+                    # linePitch: 每行的高度（用于计算行数）
+                    # charSpace: 字符间距
+                    line_pitch = docGrid.get(qn('w:linePitch'))
+                    char_space = docGrid.get(qn('w:charSpace'))
+                    
+                    # 注意：Word UI 显示的"每页行数"对应 linePitch
+                    # "每行字符数"对应 charSpace
+                    if line_pitch:
+                        lines_per_page = int(line_pitch)
+                    if char_space:
+                        chars_per_line = int(char_space)
+        except Exception as e:
+            # 如果提取网格失败，继续（网格不是必需的）
+            pass
+        
+        # 推断纸张大小
+        paper_size_inferred = infer_paper_size(page_width, page_height)
+        
+        section_data = {
+            # 页边距（twips）
+            "top_margin": top_margin,
+            "bottom_margin": bottom_margin,
+            "left_margin": left_margin,
+            "right_margin": right_margin,
+            "gutter": gutter,
+            
+            # 纸张（twips）
+            "page_width": page_width,
+            "page_height": page_height,
+            "orientation": orientation,
+            
+            # 版式（twips）
+            "header_distance": header_distance,
+            "footer_distance": footer_distance,
+            "different_first_page": different_first_page,
+            
+            # 文档网格（可选）
+            "grid_type": grid_type,
+            "chars_per_line": chars_per_line,
+            "lines_per_page": lines_per_page,
+            
+            # 推断信息
+            "paper_size_inferred": paper_size_inferred,
+        }
+        
+        sections_data.append(section_data)
+        
+        return {
+            "sections": sections_data
+        }
+        
+    except Exception as e:
+        print(f"警告: 提取页面设置失败: {e}")
+        import traceback
+        traceback.print_exc()
+        return {
+            "sections": []
+        }
+
+
 def main():
     print(f"读取文件: {DOC_PATH}")
     if not DOC_PATH.exists():
@@ -194,12 +348,16 @@ def main():
     
     # 提取编号格式
     numbering_data = extract_numbering(DOC_PATH)
+    
+    # 提取页面设置
+    page_setup_data = extract_page_setup(DOC_PATH)
 
     result = {
         "source_file": DOC_PATH.name,
         "total_styles": len(styles_data),
         "styles": styles_data,
-        "numbering": numbering_data,  # 新增：编号格式定义
+        "numbering": numbering_data,  # 编号格式定义
+        "page_setup": page_setup_data,  # 页面设置
     }
 
     OUTPUT_PATH.write_text(
@@ -212,9 +370,21 @@ def main():
         print(f"✅ 已提取编号格式定义")
     else:
         print(f"ℹ️  文档中没有编号格式")
+    
+    # 打印页面设置摘要
+    if page_setup_data and page_setup_data.get("sections"):
+        section = page_setup_data["sections"][0]
+        print(f"✅ 已提取页面设置:")
+        print(f"   - 纸张: {section.get('paper_size_inferred')} ({section.get('orientation')})")
+        print(f"   - 页边距: 上{section.get('top_margin')} 下{section.get('bottom_margin')} "
+              f"左{section.get('left_margin')} 右{section.get('right_margin')} twips")
+        if section.get('grid_type'):
+            print(f"   - 文档网格: {section.get('grid_type')} "
+                  f"(每行{section.get('chars_per_line')}字符, 每页{section.get('lines_per_page')}行)")
+    
     print(f"完整定义已保存至: {OUTPUT_PATH}")
 
-    # 打印摘要
+    # 打印样式类型分布
     by_type: dict[str, list[str]] = {}
     for s in styles_data:
         t = s["type"]