Update

RainPPR · RainPPR · commit 5b2708ed65ba · 2026-03-04T21:49:24.000+08:00
diff --git a/images/exporter-build/scripts/processor.py b/images/exporter-build/scripts/processor.py
@@ -23,27 +23,25 @@ def __init__(self, book_json_path, output_dir="build"):
     def extract_precise_toc(self, doc, offset):
         """
         根据 get_toc() 返回的初步目录，在对应页码进行文本定位，获取 y 坐标并偏移。
+        调整层级以适应整体书籍结构 (Headings 设为 Level 3+)。
         """
-        # PyMuPDF get_toc() 可能返回 3 或 4 个元素的列表: [lvl, title, page, (dest_dict)]
         raw_toc = doc.get_toc()
         refined_toc = []
         
         for entry in raw_toc:
             lvl = entry[0]
             title = entry[1]
-            page_1 = entry[2] # 1st-based page in current doc
+            page_1 = entry[2] 
             
-            # 默认目标 (整页跳转)
-            # PyMuPDF set_toc 期待 dest 为字典，或者 None (默认跳转到页顶)
+            # 原始 PDF 的 H1 (lvl 1) 在合集中应设为 Level 3
+            new_lvl = lvl + 2
             new_page_1 = page_1 + offset
             dest = {"kind": fitz.LINK_GOTO, "page": new_page_1 - 1, "to": fitz.Point(0, 0)}
             
-            # 尝试在特定页面查找标题以获取精确 Y 坐标
             page_0 = page_1 - 1
             if 0 <= page_0 < len(doc):
                 found_y = None
                 page_obj = doc[page_0]
-                # get_text("dict") 包含了文本块的边界框
                 blocks = page_obj.get_text("dict")["blocks"]
                 target_title_norm = title.strip().lower()
                 
@@ -52,75 +50,124 @@ def extract_precise_toc(self, doc, offset):
                         for line in b["lines"]:
                             for s in line["spans"]:
                                 if s["text"].strip().lower() == target_title_norm:
-                                    found_y = s["bbox"][1] # y0 (top coordinate)
+                                    found_y = s["bbox"][1]
                                     break
                             if found_y is not None: break
                     if found_y is not None: break
                 
                 if found_y is not None:
                     dest["to"] = fitz.Point(0, found_y)
-                else:
-                    print(f"    Note: Could not find precise position for '{title}' on page {page_1}, using page top.")
             
-            refined_toc.append([lvl, title, new_page_1, dest])
+            refined_toc.append([new_lvl, title, new_page_1, dest])
             
         return refined_toc
 
+    def get_english_filename(self):
+        """从 nav.json 中查找对应的英文文件名"""
+        nav_path = Path("D:/Github/blog/whk/config/nav.json")
+        if nav_path.exists():
+            try:
+                with open(nav_path, "r", encoding="utf-8") as f:
+                    nav_data = json.load(f)
+                    for item in nav_data:
+                        if item.get("title") == self.book_data.get("title"):
+                            return item.get("export", {}).get("filename", f"{self.book_data['title']}.pdf")
+            except Exception as e:
+                print(f"Error reading nav.json: {e}")
+        return f"{self.book_data['title']}.pdf"
+
     def process(self):
-        print(f"Processing Book: {self.book_data['title']}")
+        book_title = self.book_data['title']
+        print(f"Processing Book: {book_title}")
+        temp_files = []
         
-        # 1. 准备装饰页 (TeX)
-        # TODO: 渲染模板并调用 xelatex (此处假设已有编译好的 PDF 或通过外部步骤完成)
+        # 1. 插入封面与装饰页
+        decorative_pages = [
+            ("cover", f"{book_title}_cover.pdf", "封面"),
+            ("frontispiece", f"{book_title}_frontispiece.pdf", "扉页"),
+            ("toc", f"{book_title}_toc.pdf", "目录")
+        ]
         
-        # 2. 合成逻辑
-        # 插入封面 (假设名称为 cover.pdf)
-        cover_path = self.output_dir / f"{self.book_data['title']}_cover.pdf"
-        if cover_path.exists():
-            cover_doc = fitz.open(cover_path)
-            self.final_doc.insert_pdf(cover_doc)
-            self.page_offset += len(cover_doc)
-            cover_doc.close()
+        for key, fname, label in decorative_pages:
+            p = self.output_dir / fname
+            if p.exists():
+                doc = fitz.open(p)
+                self.final_doc.insert_pdf(doc)
+                self.page_offset += len(doc)
+                self.toc_data.append([1, label, self.page_offset - len(doc) + 1])
+                doc.close()
+                temp_files.append(p)
 
-        # 遍历章节
+        # 2. 遍历章节
         for section in self.book_data["sections"]:
-            print(f"  Inserting Section: {section['title']}")
+            sec_title = section['title']
+            print(f"  Inserting Section: {sec_title}")
             
-            # 插入章首页 (TeX 产物)
-            opener_path = self.output_dir / f"opener_{section['title']}.pdf"
+            # 章节首页 (Level 1)
+            opener_path = self.output_dir / f"opener_{sec_title}.pdf"
             if opener_path.exists():
                 opener_doc = fitz.open(opener_path)
                 self.final_doc.insert_pdf(opener_doc)
                 self.page_offset += len(opener_doc)
+                self.toc_data.append([1, sec_title, self.page_offset - len(opener_doc) + 1])
                 opener_doc.close()
-                self.toc_data.append([1, section['title'], self.page_offset]) # 章级目录
+                temp_files.append(opener_path)
+            else:
+                self.toc_data.append([1, sec_title, self.page_offset + 1])
 
-            # 插入内容页
+            # 插入内容页 (Level 2)
             for sub in section["sections"]:
-                # 尝试从 JSON 所在目录查找，或者使用绝对 site/build 路径
+                sub_title = sub['title']
                 content_path = self.book_json_path.parent / sub["path"]
                 if not content_path.exists():
                      content_path = Path("site/build") / sub["path"]
                      
                 if content_path.exists():
                     doc = fitz.open(content_path)
-                    # 提取并偏移章节内的书签
-                    chapter_toc = self.extract_precise_toc(doc, self.page_offset)
-                    self.toc_data.extend(chapter_toc)
+                    # 记录页面标题作为 Level 2 书签
+                    self.toc_data.append([2, sub_title, self.page_offset + 1])
+                    
+                    # 提取并偏移章节内的 headings (Level 3+)
+                    chapter_headings = self.extract_precise_toc(doc, self.page_offset)
+                    self.toc_data.extend(chapter_headings)
                     
                     self.final_doc.insert_pdf(doc)
                     self.page_offset += len(doc)
                     doc.close()
+                    temp_files.append(content_path)
                 else:
                     print(f"    Warning: Content not found at {content_path}")
 
-        # 3. 设置最终目录
+        # 3. 封底
+        back_path = self.output_dir / f"{book_title}_backcover.pdf"
+        if back_path.exists():
+            doc = fitz.open(back_path)
+            self.final_doc.insert_pdf(doc)
+            self.page_offset += len(doc)
+            self.toc_data.append([1, "封底", self.page_offset - len(doc) + 1])
+            doc.close()
+            temp_files.append(back_path)
+
+        # 4. 设置最终目录并保存
         self.final_doc.set_toc(self.toc_data)
         
-        # 4. 保存
-        output_file = self.output_dir / f"{self.book_data['title']}.pdf"
+        final_filename = self.get_english_filename()
+        output_file = self.output_dir / final_filename
         self.final_doc.save(output_file, deflate=True, garbage=4)
         self.final_doc.close()
+        
         print(f"Final PDF saved to {output_file}")
+        
+        # 5. 清理
+        print("Cleaning up temporary files...")
+        for f in temp_files:
+            try:
+                if f.exists() and f != output_file:
+                    f.unlink()
+            except: pass
+        # 清理 tex 文件
+        for f in self.output_dir.glob("*.tex"): f.unlink()
+        if (self.output_dir / "tex_tasks.txt").exists(): (self.output_dir / "tex_tasks.txt").unlink()
 
 if __name__ == "__main__":
     import argparse
@@ -135,31 +182,52 @@ def process(self):
         # 仅生成 TeX 模板供后续容器编译
         print("Rendering TeX templates...")
         generated_tex_files = []
+        book_title = processor.book_data.get('title', 'Unknown')
         
-        # 渲染封面
-        cover_filename = f"{processor.book_data.get('title', 'Unknown')}_cover.tex"
-        cover_tex = processor.jinja_env.get_template("cover.tex.j2").render(
-            title=processor.book_data.get("title", "Unknown"),
-            subtitle=processor.book_data.get("subtitle", ""),
-            authors=processor.book_data.get("authors", [])
-        )
-        cover_path = processor.output_dir / cover_filename
-        with open(cover_path, "w", encoding="utf-8") as f:
-            f.write(cover_tex)
+        common_data = {
+            "title": book_title,
+            "subtitle": processor.book_data.get("subtitle", ""),
+            "authors": processor.book_data.get("authors", []),
+            "info": processor.book_data.get("info", {})
+        }
+
+        # 1. 封面
+        cover_tex = processor.jinja_env.get_template("cover.tex.j2").render(**common_data)
+        cover_path = processor.output_dir / f"{book_title}_cover.tex"
+        with open(cover_path, "w", encoding="utf-8") as f: f.write(cover_tex)
         generated_tex_files.append(str(cover_path))
             
-        # 渲染章首页
+        # 2. 扉页
+        front_tex = processor.jinja_env.get_template("frontispiece.tex.j2").render(**common_data)
+        front_path = processor.output_dir / f"{book_title}_frontispiece.tex"
+        with open(front_path, "w", encoding="utf-8") as f: f.write(front_tex)
+        generated_tex_files.append(str(front_path))
+
+        # 3. 目录页 (简版概要)
+        toc_outline = []
+        for sec in processor.book_data["sections"]:
+            toc_outline.append({"title": sec['title'], "page": "?"}) # 物理页码在 plan 阶段未知，通常填 ? 或略过
+        toc_tex = processor.jinja_env.get_template("toc.tex.j2").render(toc_outline=toc_outline, **common_data)
+        toc_path = processor.output_dir / f"{book_title}_toc.tex"
+        with open(toc_path, "w", encoding="utf-8") as f: f.write(toc_tex)
+        generated_tex_files.append(str(toc_path))
+
+        # 4. 章首页
         for idx, section in enumerate(processor.book_data["sections"], 1):
-            opener_filename = f"opener_{section['title']}.tex"
             opener_tex = processor.jinja_env.get_template("opener.tex.j2").render(
                 chapter_num=idx,
                 chapter_title=section["title"]
             )
-            opener_path = processor.output_dir / opener_filename
-            with open(opener_path, "w", encoding="utf-8") as f:
-                f.write(opener_tex)
+            opener_path = processor.output_dir / f"opener_{section['title']}.tex"
+            with open(opener_path, "w", encoding="utf-8") as f: f.write(opener_tex)
             generated_tex_files.append(str(opener_path))
             
+        # 5. 封底
+        back_tex = processor.jinja_env.get_template("backcover.tex.j2").render(**common_data)
+        back_path = processor.output_dir / f"{book_title}_backcover.tex"
+        with open(back_path, "w", encoding="utf-8") as f: f.write(back_tex)
+        generated_tex_files.append(str(back_path))
+
         # 写入任务列表供 CI 循环调用
         with open(processor.output_dir / "tex_tasks.txt", "w", encoding="utf-8") as f:
             for tf in generated_tex_files:
diff --git a/images/exporter-build/templates/backcover.tex.j2 b/images/exporter-build/templates/backcover.tex.j2
@@ -0,0 +1,18 @@
+\documentclass[12pt,a4paper]{article}
+\usepackage[utf8]{inputenc}
+\usepackage{fontspec}
+\usepackage{geometry}
+\geometry{a4paper, margin=1in}
+
+\setmainfont{Noto Serif CJK SC}
+
+\begin{document}
+\thispagestyle{empty}
+\vspace*{15cm}
+\begin{center}
+    {\large 版权所有 \textcopyright\ \the\year\ {{ authors | join(', ') }}} \par
+    \vspace{0.5cm}
+    {\small 采用 MkDocs Exporter 构建}
+\end{center}
+\newpage
+\end{document}
diff --git a/images/exporter-build/templates/cover.tex.j2 b/images/exporter-build/templates/cover.tex.j2
@@ -4,7 +4,7 @@
 \usepackage{geometry}
 \geometry{a4paper, margin=1in}
 
-\setmainfont{Noto Sans CJK SC}
+\setmainfont{Noto Serif CJK SC}
 
 \begin{document}
 \begin{titlepage}
diff --git a/images/exporter-build/templates/frontispiece.tex.j2 b/images/exporter-build/templates/frontispiece.tex.j2
@@ -0,0 +1,31 @@
+\documentclass[12pt,a4paper]{article}
+\usepackage[utf8]{inputenc}
+\usepackage{fontspec}
+\usepackage{geometry}
+\geometry{a4paper, margin=1in}
+
+\setmainfont{Noto Serif CJK SC}
+
+\begin{document}
+\thispagestyle{empty}
+\vspace*{3cm}
+\begin{center}
+    {\Huge \bfseries {{ title }}} \par
+    \vspace{1cm}
+    {\Large {{ subtitle }}} \par
+    \vspace{2cm}
+    {\large 作者：{{ authors | join(', ') }}} \par
+    \vfill
+    \begin{minipage}{0.8\textwidth}
+        {\itshape
+        {% for line in info.abstract %}
+            {{ line }} \par \vspace{0.5cm}
+        {% endfor %}
+        }
+    \end{minipage}
+    \vfill
+    {\large {{ info.publishing }}} \par
+    {\large \today}
+\end{center}
+\newpage
+\end{document}
diff --git a/images/exporter-build/templates/opener.tex.j2 b/images/exporter-build/templates/opener.tex.j2
@@ -4,7 +4,7 @@
 \usepackage{geometry}
 \geometry{a4paper, margin=1in}
 
-\setmainfont{Noto Sans CJK SC}
+\setmainfont{Noto Serif CJK SC}
 
 \begin{document}
 \vspace*{10cm}
diff --git a/images/exporter-build/templates/toc.tex.j2 b/images/exporter-build/templates/toc.tex.j2
@@ -0,0 +1,28 @@
+\documentclass[12pt,a4paper]{article}
+\usepackage[utf8]{inputenc}
+\usepackage{fontspec}
+\usepackage{geometry}
+\geometry{a4paper, margin=1in}
+
+\setmainfont{Noto Serif CJK SC}
+
+\begin{document}
+\thispagestyle{empty}
+\section*{目录}
+\vspace{1cm}
+% 这里只是一个占位符，实际目录逻辑由 Python 处理器控制书签
+% 或者在此渲染更精美的目录
+\begin{enumerate}
+{% for section in toc_outline %}
+    \item {{ section.title }} \dotfill {{ section.page }}
+    {% if section.children %}
+    \begin{itemize}
+        {% for child in section.children %}
+            \item {{ child.title }} \dotfill {{ child.page }}
+        {% endfor %}
+    \end{itemize}
+    {% endif %}
+{% endfor %}
+\end{enumerate}
+\newpage
+\end{document}
