[Python] 纯文本查看 复制代码
"""
PDF Outline Editor
新增功能：行加粗、状态栏、折叠额外条目、保存按钮、快捷键等。
"""
import tkinter as tk
from tkinter import filedialog, messagebox, ttk
from PIL import Image, ImageTk
import fitz
import re
import pathlib
import os
try:
from tkinterdnd2 import DND_FILES, TkinterDnD
DND_AVAILABLE = True
except ImportError:
DND_AVAILABLE = False
try:
from ctypes import windll
windll.shcore.SetProcessDpiAwareness(2)
except Exception:
pass
# ---------------- 设置 ----------------
DEFAULT_COLLAPSE = 2 # 默认折叠 的 beyond 层级 (2:第一层，第二层，默认展开)，0 / None 为全展开
SET_PAEG_MODE = "UseOutlines" # 打开时显示大纲面板, 其他可选：UseNone, UseThumbs, FullScreen, UseAttachments, UseOC
# ---------------- 常量 ----------------
ICON_ADD = "img/add.png"
ICON_DEL = "img/delete.png"
ICON_INDENT = "img/indent.png"
ICON_UNINDENT = "img/unindent.png"
LLM_PROMPT = """帮我提取下目录信息，output in the following format
## The Format example 解析示例:
```
# Introduction 1
## Methods 10
### Methods_one 12
## Experiments 20
### Experiments_one 22
# End 30
```"""
TERMINOLOGY = """【术语解释】
目录页码：从OCR或原始目录中直接读取的页码，可能与PDF实际页数不符。
实际页码：PDF文件中真实的页面编号（从1开始），用于最终写入大纲。
偏移(Offset) = 实际页码 - 目录页码，通常为固定值。
额外条目：不在原目录体系中的自定义条目，可自由指定实际页码。"""
class OutlineEntry:
def __init__(self, level, title, page, is_extra=False):
self.level = level
self.title = title
self.page = page
self.actual_page = page
self.is_extra = is_extra
self.calc_offset = None
class OffsetManager:
def __init__(self):
self.mappings = []
self.extra_entries = [] # 每项: (title, actual_page, position) position="head"或"tail"
def add_mapping(self, cat_page, actual_page):
self.mappings.append((cat_page, actual_page))
def remove_mapping(self, index):
if 0 <= index < len(self.mappings):
del self.mappings[index]
def add_extra(self, title, actual_page, position="head"):
self.extra_entries.append((title, actual_page, position))
def remove_extra(self, index):
if 0 <= index < len(self.extra_entries):
del self.extra_entries[index]
def get_offset(self):
if not self.mappings:
return 0
offsets = [actual - cat for cat, actual in self.mappings]
if len(set(offsets)) == 1:
return offsets[0]
else:
messagebox.showwarning("偏移不一致", "多个对照对的页码差值不一致，请检查后重新输入。")
return None
def apply_offset(self, entries):
offset = self.get_offset()
if offset is None:
return False
for entry in entries:
if not entry.is_extra:
entry.actual_page = entry.page + offset
return True
class PDFOutlineEditor:
def __init__(self, root):
self.root = root
self.current_file = None
self.max_page_number = 0
self.entries = []
self.edit_mode = "indent"
self.offset_mgr = OffsetManager()
self.use_icons = all(os.path.exists(ic) for ic in [ICON_ADD, ICON_DEL, ICON_INDENT, ICON_UNINDENT])
if self.use_icons:
self.icons = {
'add': ImageTk.PhotoImage(Image.open(ICON_ADD).resize((20,20))),
'del': ImageTk.PhotoImage(Image.open(ICON_DEL).resize((20,20))),
'indent': ImageTk.PhotoImage(Image.open(ICON_INDENT).resize((20,20))),
'unindent': ImageTk.PhotoImage(Image.open(ICON_UNINDENT).resize((20,20)))
}
self.create_widgets()
self.setup_dnd()
self.bind_shortcuts()
def create_widgets(self):
main_pane = tk.PanedWindow(self.root, orient=tk.HORIZONTAL)
main_pane.pack(fill=tk.BOTH, expand=1)
left_frame = tk.Frame(main_pane)
self.build_file_panel(left_frame)
self.build_offset_panel(left_frame)
main_pane.add(left_frame)
right_frame = tk.Frame(main_pane)
self.build_editor_panel(right_frame)
main_pane.add(right_frame)
# 底部状态栏
self.status_bar = tk.Label(self.root, text="总页数：未加载", bd=1, relief=tk.SUNKEN, anchor=tk.W)
self.status_bar.pack(side=tk.BOTTOM, fill=tk.X)
self.build_menu()
def build_menu(self):
menubar = tk.Menu(self.root)
self.root.config(menu=menubar)
file_menu = tk.Menu(menubar, tearoff=0)
file_menu.add_command(label="选择PDF", command=self.load_pdf)
file_menu.add_command(label="保存到PDF", command=self.save_pdf, accelerator="Ctrl+S")
file_menu.add_separator()
file_menu.add_command(label="退出", command=self.root.quit)
menubar.add_cascade(label="文件", menu=file_menu)
import_menu = tk.Menu(menubar, tearoff=0)
import_menu.add_command(label="导入现成OCR版本...", command=self.open_ocr_subwin)
menubar.add_cascade(label="导入", menu=import_menu)
help_menu = tk.Menu(menubar, tearoff=0)
help_menu.add_command(label="显示LLM OCR提示词", command=self.show_prompt)
help_menu.add_separator()
help_menu.add_command(label="软件术语解释", command=self.show_terminology)
menubar.add_cascade(label="帮助", menu=help_menu)
def build_file_panel(self, parent):
frame = tk.LabelFrame(parent, text="文件操作")
tk.Button(frame, text="选择PDF", command=self.load_pdf).pack(pady=5, padx=5)
self.file_label = tk.Label(frame, text="未选择文件", fg="gray")
self.file_label.pack()
if DND_AVAILABLE:
self.drop_target = tk.Label(frame, text="拖放PDF文件至此", relief="sunken")
self.drop_target.pack(pady=10, ipadx=20, ipady=10)
frame.pack(pady=10, padx=5, fill=tk.X)
def build_offset_panel(self, parent):
frame = tk.LabelFrame(parent, text="页码补偿 & 预览", padx=12, pady=12)
# 对照表区域
map_frame = tk.Frame(frame)
tk.Label(map_frame, text="目录页码 -> 实际页码").grid(row=0, column=0, columnspan=2)
self.map_listbox = tk.Listbox(map_frame, height=4, width=30)
self.map_listbox.grid(row=1, column=0, rowspan=3, sticky='nsew')
scrollbar = tk.Scrollbar(map_frame, orient=tk.VERTICAL, command=self.map_listbox.yview)
scrollbar.grid(row=1, column=1, rowspan=3, sticky='ns')
self.map_listbox.config(yscrollcommand=scrollbar.set)
tk.Label(map_frame, text="目录页码:").grid(row=1, column=2, sticky='w')
self.cat_page_entry = tk.Entry(map_frame, width=10)
self.cat_page_entry.grid(row=1, column=3)
tk.Label(map_frame, text="实际页码:").grid(row=2, column=2, sticky='w')
self.actual_page_entry = tk.Entry(map_frame, width=10)
self.actual_page_entry.grid(row=2, column=3)
tk.Button(map_frame, text="添加对照", command=self.add_mapping).grid(row=3, column=2, columnspan=2)
tk.Button(map_frame, text="删除选中", command=self.del_mapping).grid(row=4, column=2, columnspan=2)
map_frame.pack(pady=5)
# 额外条目区域（可折叠）
self.extra_visible = False
self.extra_frame_container = tk.Frame(frame)
self.toggle_btn = tk.Button(frame, text="展开额外条目 ▸", command=self.toggle_extra_panel)
self.toggle_btn.pack(pady=5, anchor='w')
extra_frame = tk.Frame(self.extra_frame_container)
tk.Label(extra_frame, text="额外条目（不在目录体系中）").grid(row=0, column=0, columnspan=2)
self.extra_listbox = tk.Listbox(extra_frame, height=3, width=30)
self.extra_listbox.grid(row=1, column=0, rowspan=4, sticky='nsew')
scrollbar2 = tk.Scrollbar(extra_frame, orient=tk.VERTICAL, command=self.extra_listbox.yview)
scrollbar2.grid(row=1, column=1, rowspan=4, sticky='ns')
self.extra_listbox.config(yscrollcommand=scrollbar2.set)
tk.Label(extra_frame, text="标题:").grid(row=1, column=2, sticky='w')
self.extra_title_entry = tk.Entry(extra_frame, width=10)
self.extra_title_entry.grid(row=1, column=3)
tk.Label(extra_frame, text="实际页码:").grid(row=2, column=2, sticky='w')
self.extra_page_entry = tk.Entry(extra_frame, width=10)
self.extra_page_entry.grid(row=2, column=3)
tk.Label(extra_frame, text="位置:").grid(row=3, column=2, sticky='w')
self.extra_pos_var = tk.StringVar(value="head")
tk.OptionMenu(extra_frame, self.extra_pos_var, "head", "tail").grid(row=3, column=3)
tk.Button(extra_frame, text="添加条目", command=self.add_extra_entry).grid(row=4, column=2, columnspan=2)
tk.Button(extra_frame, text="删除选中", command=self.del_extra_entry).grid(row=5, column=2, columnspan=2)
extra_frame.pack(pady=5)
frame.pack(pady=10, padx=5, fill=tk.X)
# 预览按钮
tk.Button(parent, text="预览前三条目截图", command=self.preview_pages).pack(pady=5)
def toggle_extra_panel(self):
if self.extra_visible:
self.extra_frame_container.pack_forget()
self.toggle_btn.config(text="展开额外条目 ▸")
else:
self.extra_frame_container.pack(pady=5)
self.toggle_btn.config(text="折叠额外条目 ▾")
self.extra_visible = not self.extra_visible
def build_editor_panel(self, parent):
frame = tk.Frame(parent)
toolbar = tk.Frame(frame)
self.create_tool_button(toolbar, 'add', "添加条目", self.add_entry)
self.create_tool_button(toolbar, 'del', "删除条目", self.delete_entry)
self.create_tool_button(toolbar, 'indent', "增加缩进", self.indent_entry)
self.create_tool_button(toolbar, 'unindent', "减少缩进", self.unindent_entry)
self.mode_btn = tk.Button(toolbar, text="Markdown版", command=self.toggle_mode)
self.mode_btn.pack(side=tk.RIGHT, padx=5)
toolbar.pack(fill=tk.X, pady=2)
# 带滚动条的文本编辑器
text_frame = tk.Frame(frame)
text_frame.pack(fill=tk.BOTH, expand=1, padx=5, pady=5)
self.text_editor = tk.Text(text_frame, wrap=tk.NONE, undo=True, width=50, height=35)
scroll_y = tk.Scrollbar(text_frame, orient=tk.VERTICAL, command=self.text_editor.yview)
self.text_editor.configure(yscrollcommand=scroll_y.set)
self.text_editor.grid(row=0, column=0, sticky='nsew')
scroll_y.grid(row=0, column=1, sticky='ns')
text_frame.grid_rowconfigure(0, weight=1)
text_frame.grid_columnconfigure(0, weight=1)
# 加粗tag
self.text_editor.tag_configure("bold", font=("TkDefaultFont", 10, "bold"))
self.text_editor.bind("<Button-1>", self.on_click_select_line)
self.text_editor.bind("<Double-1>", self.on_double_click_edit)
# 右侧原有保存按钮保留（也可用）
tk.Button(frame, text="保存到PDF", command=self.save_pdf).pack(pady=5)
frame.pack(fill=tk.BOTH, expand=1)
def create_tool_button(self, parent, icon_key, text, command):
if self.use_icons:
btn = tk.Button(parent, image=self.icons[icon_key], command=command)
else:
btn = tk.Button(parent, text=text, command=command)
btn.pack(side=tk.LEFT, padx=2, pady=2)
def setup_dnd(self):
if DND_AVAILABLE:
try:
self.root.drop_target_register(DND_FILES)
self.root.dnd_bind('<<Drop>>', self.handle_drop)
except Exception:
pass
def bind_shortcuts(self):
self.root.bind('<Tab>', lambda e: self.indent_entry())
self.root.bind('<Shift-Tab>', lambda e: self.unindent_entry())
self.root.bind('<Control-s>', lambda e: self.save_pdf())
self.text_editor.bind('<Tab>', self.focus_indent)
self.text_editor.bind('<Shift-Tab>', self.focus_unindent)
def focus_indent(self, event):
self.indent_entry()
return "break"
def focus_unindent(self, event):
self.unindent_entry()
return "break"
def handle_drop(self, event):
filepath = event.data.strip('{}')
if filepath.lower().endswith('.pdf'):
self.load_pdf(filepath)
def load_pdf(self, path=None):
if not path:
path = filedialog.askopenfilename(filetypes=[("PDF文件", "*.pdf")])
if not path:
return
try:
with fitz.open(path) as doc:
self.max_page_number = doc.page_count
toc = doc.get_toc()
self.entries = []
for level, title, page in toc:
self.entries.append(OutlineEntry(level, title, page))
self.current_file = path
self.file_label.config(text=os.path.basename(path), fg='blue')
self.status_bar.config(text=f"总页数：{self.max_page_number}")
# 清理 offset 管理器
self.offset_mgr = OffsetManager()
self.refresh_map_listbox()
self.refresh_extra_listbox()
self.cat_page_entry.delete(0, tk.END)
self.actual_page_entry.delete(0, tk.END)
self.extra_title_entry.delete(0, tk.END)
self.extra_page_entry.delete(0, tk.END)
self.refresh_display()
except Exception as e:
messagebox.showerror("错误", f"无法读取PDF:\n{str(e)}")
def save_pdf(self):
# 确保最新编辑内容
self.parse_display()
if not self.current_file:
messagebox.showwarning("提示", "请先选择PDF文件")
return
if not self.entries:
messagebox.showwarning("提示", "大纲为空")
return
if not self.offset_mgr.apply_offset(self.entries):
return
if not self.validate_hierarchy():
return
toc = []
# 先处理位置为"head"的额外条目
head_extras = [e for e in self.offset_mgr.extra_entries if e[2] == "head"]
tail_extras = [e for e in self.offset_mgr.extra_entries if e[2] == "tail"]
for title, page, _ in head_extras:
toc.append([1, title, page])
for entry in self.entries:
page = entry.actual_page
if page > self.max_page_number:
page = self.max_page_number
elif page < 1:
page = 1
toc.append([entry.level, entry.title, page])
for title, page, _ in tail_extras:
toc.append([1, title, page])
file_path = pathlib.Path(self.current_file)
save_path = filedialog.asksaveasfilename(
defaultextension=".pdf",
initialdir=file_path.parent,
initialfile=file_path.name,
filetypes=[("PDF文件", "*.pdf")]
)
if not save_path:
return
try:
with fitz.open(self.current_file) as doc:
doc.set_toc(toc,collapse=DEFAULT_COLLAPSE) #
# 打开时显示大纲面板
doc.set_pagemode(SET_PAEG_MODE)
doc.save(save_path)
messagebox.showinfo("成功", "文件保存成功")
except Exception as e:
messagebox.showerror("错误", f"保存失败:\n{str(e)}")
def validate_hierarchy(self):
levels = [entry.level for entry in self.entries]
problems = []
prev = 0
for i, lvl in enumerate(levels):
if lvl - prev > 1:
problems.append(f"第{i+1}行：从{prev}级跳至{lvl}级")
prev = lvl
if problems:
messagebox.showwarning("层级跳跃警告",
"发现以下层级跳跃，可能导致PDF大纲显示异常：\n" + "\n".join(problems))
return True
def refresh_display(self):
self.text_editor.delete(1.0, tk.END)
if self.edit_mode == "indent":
text = self.entries_to_indent()
else:
text = self.entries_to_markdown()
self.text_editor.insert(1.0, text)
# 对一级标题加粗
self.apply_bold_tags()
def parse_display(self):
raw = self.text_editor.get(1.0, tk.END)
if self.edit_mode == "indent":
self.entries = self.parse_indent_text(raw)
else:
self.entries = self.parse_markdown_text(raw)
def entries_to_indent(self):
lines = []
for e in self.entries:
indent = " " * (e.level - 1)
lines.append(f"{indent}{e.title} ({e.page})")
return "\n".join(lines)
def entries_to_markdown(self):
lines = []
for e in self.entries:
prefix = "#" * e.level
lines.append(f"{prefix} {e.title} {e.page}")
return "\n".join(lines)
def parse_indent_text(self, text):
entries = []
for line in text.splitlines():
line = line.rstrip()
if not line:
continue
match = re.match(r'^(\s*)\S', line)
if not match:
continue
spaces = len(match.group(1))
level = spaces // 2 + 1
content = line.lstrip()
m = re.match(r'^(.+)\s+\((\d+)\)$', content)
if not m:
m = re.match(r'^(.+)\s+(\d+)$', content)
if m:
title, page = m.group(1).strip(), int(m.group(2))
entries.append(OutlineEntry(level, title, page))
return entries
def parse_markdown_text(self, text):
entries = []
for line in text.splitlines():
line = line.strip()
if not line or not line.startswith('#'):
continue
m = re.match(r'^(#+)\s+(.+)\s+(\d+)$', line)
if m:
level = len(m.group(1))
title = m.group(2).strip()
page = int(m.group(3))
entries.append(OutlineEntry(level, title, page))
return entries
def apply_bold_tags(self):
self.text_editor.tag_remove("bold", "1.0", tk.END)
for line_num in range(1, int(self.text_editor.index('end').split('.')[0])):
line_text = self.text_editor.get(f"{line_num}.0", f"{line_num}.end")
if self.edit_mode == "indent":
# 无缩进或以0个空格开头的（一级）
if not line_text.startswith(" "):
self.text_editor.tag_add("bold", f"{line_num}.0", f"{line_num}.end")
else:
if line_text.startswith("# ") and not line_text.startswith("##"):
self.text_editor.tag_add("bold", f"{line_num}.0", f"{line_num}.end")
def toggle_mode(self):
self.parse_display()
if self.edit_mode == "indent":
self.edit_mode = "markdown"
self.mode_btn.config(text="缩进版")
else:
self.edit_mode = "indent"
self.mode_btn.config(text="Markdown版")
self.refresh_display()
def add_entry(self):
self.text_editor.insert(tk.END, "\n新条目 (1)")
self.parse_display()
def delete_entry(self):
try:
sel = self.text_editor.tag_ranges("sel")
if sel:
self.text_editor.delete(sel[0], sel[1])
self.parse_display()
else:
messagebox.showinfo("提示", "请先点击行选择")
except Exception:
pass
def indent_entry(self):
self.adjust_indent(2)
def unindent_entry(self):
self.adjust_indent(-2)
def adjust_indent(self, delta):
try:
start_line = int(self.text_editor.index(tk.SEL_FIRST).split('.')[0])
end_line = int(self.text_editor.index(tk.SEL_LAST).split('.')[0])
except tk.TclError:
cursor_line = int(self.text_editor.index(tk.INSERT).split('.')[0])
start_line = end_line = cursor_line
for line_num in range(start_line, end_line + 1):
line_text = self.text_editor.get(f"{line_num}.0", f"{line_num}.end")
current_indent = len(line_text) - len(line_text.lstrip(' '))
new_indent = max(0, current_indent + delta)
new_text = ' ' * new_indent + line_text.lstrip(' ')
self.text_editor.replace(f"{line_num}.0", f"{line_num}.end", new_text)
self.text_editor.tag_remove("sel", "1.0", tk.END)
self.text_editor.tag_add("sel", f"{start_line}.0", f"{end_line}.end")
self.parse_display()
def on_click_select_line(self, event):
index = self.text_editor.index(f"@{event.x},{event.y}")
line_start = index.split('.')[0] + ".0"
line_end = index.split('.')[0] + ".end"
self.text_editor.tag_remove("sel", "1.0", tk.END)
self.text_editor.tag_add("sel", line_start, line_end)
def on_double_click_edit(self, event):
index = self.text_editor.index(f"@{event.x},{event.y}")
line_num = int(index.split('.')[0])
line_text = self.text_editor.get(f"{line_num}.0", f"{line_num}.end")
if self.edit_mode == "indent":
title, page = self.parse_indent_line(line_text)
level = (len(line_text) - len(line_text.lstrip(' '))) // 2 + 1
else:
title, page, level = self.parse_markdown_line(line_text)
if title is None:
return
dialog = EditEntryDialog(self.root, title, page, level, self.max_page_number)
self.root.wait_window(dialog.top)
if dialog.result:
new_title, new_page, new_level = dialog.result
if self.edit_mode == "indent":
new_line = " " * (new_level - 1) + f"{new_title} ({new_page})"
else:
new_line = "#" * new_level + f" {new_title} {new_page}"
self.text_editor.replace(f"{line_num}.0", f"{line_num}.end", new_line)
self.parse_display()
def parse_indent_line(self, text):
content = text.lstrip()
m = re.match(r'^(.+)\s+\((\d+)\)$', content)
if m:
return m.group(1).strip(), int(m.group(2))
return None, None
def parse_markdown_line(self, text):
m = re.match(r'^(#+)\s+(.+)\s+(\d+)$', text.strip())
if m:
level = len(m.group(1))
return m.group(2).strip(), int(m.group(3)), level
return None, None, None
def open_ocr_subwin(self):
sub = tk.Toplevel(self.root)
sub.title("导入现成OCR版本")
sub.geometry("500x400")
tk.Label(sub, text="粘贴OCR内容（格式：页码 (p...): 标题）").pack(pady=5)
text = tk.Text(sub, wrap=tk.WORD, width=60, height=15)
text.pack(padx=10, pady=5, fill=tk.BOTH, expand=True)
# 示例文本
EXAMPLE = (
"1 (p1): 示例\n"
"3 (p2): 第1章 标题\n"
"3 (p2-1): 第一节 标题\n"
"5 (p2-2-1): 第二节 第一子节 标题\n"
)
text.insert("1.0", EXAMPLE)
text.tag_configure("placeholder", foreground="grey")
text.tag_add("placeholder", "1.0", "end")
def on_focus_in(event):
if text.get("1.0", "end-1c").strip() == EXAMPLE.strip():
text.delete("1.0", "end")
text.config(foreground="black")
def on_focus_out(event):
if not text.get("1.0", "end-1c").strip():
text.insert("1.0", EXAMPLE)
text.tag_add("placeholder", "1.0", "end")
text.bind("<FocusIn>", on_focus_in)
text.bind("<FocusOut>", on_focus_out)
def do_import():
raw = text.get(1.0, tk.END)
# 如果内容仍是示例，视为空
if raw.strip() == EXAMPLE.strip():
raw = ""
new_entries = self.parse_ocr_text(raw)
if not new_entries:
messagebox.showwarning("错误", "未能解析出有效条目")
return
self.entries = new_entries
self.refresh_display()
sub.destroy()
tk.Button(sub, text="确定导入", command=do_import).pack(pady=10)
def parse_ocr_text(self, raw):
entries = []
lines = raw.splitlines()
non_empty = 0
valid = 0
for line in lines:
line = line.strip()
if not line:
continue
non_empty += 1
# 匹配: 页码 (p层级链): 标题
m = re.match(r'(\d+)\s+\(p([\d\-]+)\):\s*(.*)', line)
if not m:
continue
valid += 1
cat_page = int(m.group(1))
p_chain = m.group(2) # 如 "2-1-1"
title = m.group(3).strip()
level = len(p_chain.split('-')) # 段数即层级
entries.append(OutlineEntry(level, title, cat_page))
if valid < non_empty:
messagebox.showwarning("解析警告",
f"输入了 {non_empty} 个非空行，仅解析出 {valid} 个条目，请检查格式。")
return entries
def add_mapping(self):
try:
cat = int(self.cat_page_entry.get())
actual = int(self.actual_page_entry.get())
except ValueError:
messagebox.showwarning("错误", "请输入有效数字")
return
self.offset_mgr.add_mapping(cat, actual)
self.refresh_map_listbox()
self.cat_page_entry.delete(0, tk.END)
self.actual_page_entry.delete(0, tk.END)
def del_mapping(self):
sel = self.map_listbox.curselection()
if sel:
self.offset_mgr.remove_mapping(sel[0])
self.refresh_map_listbox()
def refresh_map_listbox(self):
self.map_listbox.delete(0, tk.END)
for cat, actual in self.offset_mgr.mappings:
self.map_listbox.insert(tk.END, f"{cat} -> {actual}")
def add_extra_entry(self):
title = self.extra_title_entry.get().strip()
if not title:
messagebox.showwarning("错误", "请输入标题")
return
try:
page = int(self.extra_page_entry.get())
except ValueError:
messagebox.showwarning("错误", "请输入有效页码")
return
position = self.extra_pos_var.get() # "head" or "tail"
self.offset_mgr.add_extra(title, page, position)
self.refresh_extra_listbox()
self.extra_title_entry.delete(0, tk.END)
self.extra_page_entry.delete(0, tk.END)
def del_extra_entry(self):
sel = self.extra_listbox.curselection()
if sel:
self.offset_mgr.remove_extra(sel[0])
self.refresh_extra_listbox()
def refresh_extra_listbox(self):
self.extra_listbox.delete(0, tk.END)
for title, page, pos in self.offset_mgr.extra_entries:
self.extra_listbox.insert(tk.END, f"[{'从头' if pos=='head' else '从尾'}] {title} : {page}")
def preview_pages(self):
# 同步最新的编辑器内容
self.parse_display()
if not self.current_file or not self.entries:
messagebox.showwarning("提示", "请先加载PDF并确保大纲不为空")
return
target_entries = []
for level in range(1, 5):
for e in self.entries:
if e.level == level and len(target_entries) < 3:
target_entries.append(e)
if len(target_entries) >= 3:
break
if not target_entries:
messagebox.showwarning("提示", "没有有效条目")
return
preview = tk.Toplevel(self.root)
preview.title("预览前三条目对应页面")
canvas = tk.Canvas(preview, width=600, height=800)
canvas.pack(side=tk.LEFT, fill=tk.BOTH, expand=True)
scrollbar = tk.Scrollbar(preview, orient=tk.VERTICAL, command=canvas.yview)
scrollbar.pack(side=tk.RIGHT, fill=tk.Y)
canvas.configure(yscrollcommand=scrollbar.set)
inner = tk.Frame(canvas)
canvas.create_window((0,0), window=inner, anchor='nw')
try:
doc = fitz.open(self.current_file)
offset = self.offset_mgr.get_offset()
if offset is None and self.offset_mgr.mappings:
offset = 0
elif offset is None:
offset = 0
for i, entry in enumerate(target_entries):
actual = entry.page + offset if not entry.is_extra else entry.actual_page
page_idx = max(0, min(actual - 1, doc.page_count - 1))
page = doc.load_page(page_idx)
pix = page.get_pixmap(dpi=100)
img = Image.frombytes("RGB", [pix.width, pix.height], pix.samples)
img_tk = ImageTk.PhotoImage(img)
label_frame = tk.Frame(inner)
tk.Label(label_frame, text=f"条目: {entry.title}", font=('Arial', 10, 'bold')).pack()
tk.Label(label_frame, text=f"目录页码: {entry.page} 实际页码: {actual}").pack()
lbl_img = tk.Label(label_frame, image=img_tk)
lbl_img.image = img_tk
lbl_img.pack()
label_frame.pack(pady=10)
doc.close()
except Exception as e:
messagebox.showerror("错误", f"预览失败: {str(e)}")
inner.update_idletasks()
canvas.config(scrollregion=canvas.bbox("all"))
def show_prompt(self):
prompt_win = tk.Toplevel(self.root)
prompt_win.title("LLM OCR 提示词")
text = tk.Text(prompt_win, wrap=tk.WORD, width=70, height=15)
text.insert(1.0, LLM_PROMPT)
text.pack(padx=10, pady=10)
def copy_to_clipboard():
self.root.clipboard_clear()
self.root.clipboard_append(LLM_PROMPT)
# messagebox.showinfo("已复制", "提示词已复制到剪贴板")
tk.Button(prompt_win, text="复制到剪贴板", command=copy_to_clipboard).pack(pady=5)
def show_terminology(self):
term_win = tk.Toplevel(self.root)
term_win.title("软件术语解释")
text = tk.Text(term_win, wrap=tk.WORD, width=60, height=10)
text.insert(1.0, TERMINOLOGY)
text.config(state=tk.DISABLED)
text.pack(padx=10, pady=10)
class EditEntryDialog:
def __init__(self, parent, title, page, level, max_page):
self.top = tk.Toplevel(parent)
self.top.title("编辑条目")
self.result = None
tk.Label(self.top, text="标题:").grid(row=0, column=0, padx=5, pady=5, sticky='w')
self.title_var = tk.StringVar(value=title)
tk.Entry(self.top, textvariable=self.title_var, width=30).grid(row=0, column=1, padx=5, pady=5)
tk.Label(self.top, text="页码:").grid(row=1, column=0, padx=5, pady=5, sticky='w')
self.page_var = tk.IntVar(value=page)
tk.Spinbox(self.top, from_=1, to=max_page or 9999, textvariable=self.page_var, width=8).grid(row=1, column=1, padx=5, pady=5, sticky='w')
tk.Label(self.top, text="层级:").grid(row=2, column=0, padx=5, pady=5, sticky='w')
self.level_var = tk.IntVar(value=level)
tk.Spinbox(self.top, from_=1, to=5, textvariable=self.level_var, width=5).grid(row=2, column=1, padx=5, pady=5, sticky='w')
btn_frame = tk.Frame(self.top)
tk.Button(btn_frame, text="确定", command=self.on_ok).pack(side=tk.LEFT, padx=5)
tk.Button(btn_frame, text="取消", command=self.top.destroy).pack(side=tk.LEFT, padx=5)
btn_frame.grid(row=3, column=0, columnspan=2, pady=10)
def on_ok(self):
self.result = (self.title_var.get().strip(), self.page_var.get(), self.level_var.get())
self.top.destroy()
if __name__ == "__main__":
if DND_AVAILABLE:
root = TkinterDnD.Tk()
else:
root = tk.Tk()
root.title("PDF大纲编辑器 Pro")
root.geometry("900x700")
app = PDFOutlineEditor(root)
root.mainloop()
[复制链接]
发表于 2026-6-14 13:47
|
发表于 2026-6-14 15:20
|楼主
分享到朋友圈
