# -*- coding: utf-8 -*-
#!/usr/bin/env python3
"""
建立標準 EPUB 3.0
"""
import re
import os
import zipfile
def create_epub(text_file, output_file):
with open(text_file, 'r', encoding='utf-8') as f:
text = f.read()
# 分割章節
chapter_pattern = r'^(第[一二三四五六七八九十百零]+章\s+.+)$'
parts = re.split(f'({chapter_pattern})', text, flags=re.MULTILINE)
chapters = []
i = 0
while i < len(parts):
chunk = parts[i].strip()
if re.match(chapter_pattern, chunk) and i + 1 < len(parts):
title = chunk
content = parts[i + 1].strip() if i + 1 < len(parts) else ''
chapters.append((title, content))
i += 2
elif chunk and len(chunk) > 50:
chapters.append(('前言', chunk))
i += 1
else:
i += 1
if len(chapters) < 5:
paragraphs = [p.strip() for p in text.split('\n\n') if len(p.strip()) > 50]
for j in range(0, len(paragraphs), 10):
chunk = '\n\n'.join(paragraphs[j:j+10])
chapters.append((f'段落 {j//10+1}', chunk))
print(f'共 {len(chapters)} 個章節')
# 建立 EPUB zip
with zipfile.ZipFile(output_file, 'w', zipfile.ZIP_DEFLATED) as epub:
# mimetype（必須第一個，不壓縮）
epub.writestr('mimetype', 'application/epub+zip', compress_type=zipfile.ZIP_STORED)
# META-INF/container.xml
container = '''
'''
epub.writestr('META-INF/container.xml', container)
# content.opf
manifest_items = ''
spine_items = ''
toc_nav = ''
for i, (title, _) in enumerate(chapters):
manifest_items += f' {title}
行動中的組織
詹姆斯·湯普森
zh-Hant
thompson-organizations-in-action-zh-final-v2
{spine_items}
'''
epub.writestr('OEBPS/content.opf', opf)
# toc.ncx
nav_points = ''
for i, (title, _) in enumerate(chapters):
safe_title = title.replace('&', '&').replace('<', '<')
nav_points += f'''
{safe_title} \n'''
ncx = f'''
行動中的組織
{nav_points}
'''
epub.writestr('OEBPS/toc.ncx', ncx)
# nav.xhtml
nav_items = ''
for i, (title, _) in enumerate(chapters):
safe_title = title.replace('&', '&').replace('<', '<')
nav_items += f' {safe_title} 目錄
目錄
{nav_items}
'''
epub.writestr('OEBPS/nav.xhtml', nav_xhtml)
# 章節 XHTML
for i, (title, content) in enumerate(chapters):
safe_title = title.replace('&', '&').replace('<', '<')
paragraphs = content.split('\n\n')
body = f' {safe_title} \n'
for para in paragraphs:
para = para.strip()
if para:
safe_para = para.replace('&', '&').replace('<', '<')
# 保留換行
safe_para = safe_para.replace('\n', '{safe_para}
\n'
xhtml = f'''
{safe_title}
{body}
'''
epub.writestr(f'OEBPS/chapter{i}.xhtml', xhtml)
size = os.path.getsize(output_file)
print(f'EPUB: {output_file} ({size:,} bytes)')
if __name__ == '__main__':
create_epub(
'/Users/lingochen/Desktop/ocr_final_clean.txt',
'/Users/lingochen/Desktop/行動中的組織_最終版v2.epub'
)