#!/usr/bin/env python3
"""epubgen - generate spec-perfect EPUB 3.3 files for testing readers and validators.
Two modes:
normal - a clean, readable EPUB that uses the full feature set
(cover, title page, TOC, landmarks, chapters, index, colophon,
EPUB2 NCX fallback, EPUB3 nav, CSS, metadata refinements).
torture - a technically valid EPUB with deliberately hostile syntax
(bidi overrides, combining chars, numeric char refs, deep nested
markup, unusual-but-valid xml:lang, weird filenames, ...).
Every XML file is re-parsed with xml.etree before zipping, so the output is
guaranteed well-formed. Zero dependencies beyond the Python standard library.
Usage:
python3 epubgen.py [options]
"""
import argparse
import html
import os
import random
import re
import struct
import sys
import uuid
import xml.etree.ElementTree as ET
import zipfile
import zlib
from datetime import datetime, timezone
from urllib.parse import quote
VERSION = "1.0"
HERE = os.path.dirname(os.path.abspath(__file__))
DEFAULT_FILE = os.path.join(HERE, "paragraphs.txt")
NS = 'xmlns="http://www.w3.org/1999/xhtml" xmlns:epub="http://www.idpf.org/2007/ops"'
# Hostile-but-valid Unicode (all fine in XML 1.0 / UTF-8)
COMBINING = "\u0300\u0301\u0308\u030a\u20dd\u20e3"
ZW = "\u200b\u200c\u200d\u200e\u200f"
SPACES = "\u00a0\u2009\u200a\u202f\u3000"
BIDI = "\u202a\u202b\u202d\u202e"
PRIV = "\ue000\ue100\uf8ff"
CJK = "\u6c49\u5b57\u6d4b\u8bd5\u30c6\u30b9\u30c8"
EMOJI = "\U0001f600\U0001f389\U0001f680\U0001f480\U0001f518\U0001f30d"
def esc(s):
"""XML-escape text (& < > and quotes; escaping '>' makes ']]>' impossible)."""
return html.escape(s, quote=True)
def xhtml(title, body, body_type=None, link_css=True):
"""Wrap body content in a spec-compliant XHTML5 document."""
head = f'{esc(title)}'
if link_css:
head += ''
head += ''
btype = f' epub:type="{body_type}"' if body_type else ''
return ('\n'
f'{head}{body}