epub2md/tests/conftest.py

"""Builders for small synthetic EPUB files used throughout the tests."""

from __future__ import annotations

import json
import sys
import zipfile
from dataclasses import dataclass, field
from pathlib import Path

import pytest

ROOT = Path(__file__).resolve().parent.parent
sys.path.insert(0, str(ROOT))

from epub2md.cli import main  # noqa: E402

# 1x1 transparent PNG
PNG = bytes.fromhex(
    "89504e470d0a1a0a0000000d49484452000000010000000108060000001f15c489"
    "0000000d4944415478da63f8ffff3f0005fe02fea7d6a5b60000000049454e44ae426082"
)
SVG = b'<svg xmlns="http://www.w3.org/2000/svg" width="10" height="10"><rect width="10" height="10"/></svg>'

XHTML = """<?xml version="1.0" encoding="utf-8"?>
<!DOCTYPE html>
<html xmlns="http://www.w3.org/1999/xhtml" xmlns:epub="http://www.idpf.org/2007/ops" lang="de">
<head><title>{title}</title></head>
<body>{body}</body>
</html>"""


@dataclass
class Doc:
    name: str  # file name inside OEBPS/
    body: str
    title: str = ""
    linear: bool = True


@dataclass
class Book:
    docs: list[Doc]
    version: str = "3.0"
    title: str = "Testbuch"
    authors: list[str] = field(default_factory=lambda: ["Erika Musterfrau"])
    language: str = "de"
    identifier: str = "urn:isbn:9783161484100"
    media: dict[str, bytes] = field(default_factory=dict)  # OEBPS-relative path -> bytes
    nav: list[tuple[str, str]] | None = None  # (label, href) – default: one entry per doc
    ncx: bool = False
    extra_entries: dict[str, bytes] = field(default_factory=dict)  # raw ZIP entries (any name)
    container: str | None = None
    opf_override: str | None = None
    spine_extra: str = ""
    cover: str | None = None


MEDIA_TYPES = {".png": "image/png", ".jpg": "image/jpeg", ".svg": "image/svg+xml", ".gif": "image/gif",
               ".mp3": "audio/mpeg", ".css": "text/css", ".otf": "font/otf"}


def build_epub(path: Path, book: Book) -> Path:
    manifest = []
    spine = []
    for i, d in enumerate(book.docs):
        manifest.append(f'<item id="d{i}" href="{d.name}" media-type="application/xhtml+xml"/>')
        spine.append(f'<itemref idref="d{i}"' + ("" if d.linear else ' linear="no"') + "/>")
    for i, (name, _) in enumerate(book.media.items()):
        ext = "." + name.rsplit(".", 1)[-1].lower()
        props = ' properties="cover-image"' if book.cover == name else ""
        manifest.append(f'<item id="m{i}" href="{name}" media-type="{MEDIA_TYPES.get(ext, "application/octet-stream")}"{props}/>')
    nav_entries = book.nav if book.nav is not None else [(d.title or d.name, d.name) for d in book.docs]
    epub3 = book.version.startswith("3")
    if epub3:
        manifest.append('<item id="nav" href="nav.xhtml" media-type="application/xhtml+xml" properties="nav"/>')
    if book.ncx:
        manifest.append('<item id="ncx" href="toc.ncx" media-type="application/x-dtbncx+xml"/>')
    creators = "".join(f"<dc:creator>{a}</dc:creator>" for a in book.authors)
    opf = book.opf_override or f"""<?xml version="1.0" encoding="utf-8"?>
<package xmlns="http://www.idpf.org/2007/opf" version="{book.version}" unique-identifier="uid">
  <metadata xmlns:dc="http://purl.org/dc/elements/1.1/">
    <dc:title>{book.title}</dc:title>{creators}
    <dc:language>{book.language}</dc:language>
    <dc:identifier id="uid">{book.identifier}</dc:identifier>
    <dc:publisher>Testverlag</dc:publisher>
    <dc:date>2024-01-01</dc:date>
  </metadata>
  <manifest>{''.join(manifest)}</manifest>
  <spine{' toc="ncx"' if book.ncx else ''}>{''.join(spine)}{book.spine_extra}</spine>
</package>"""
    nav_xhtml = XHTML.format(
        title="Inhalt",
        body='<nav epub:type="toc"><ol>' + "".join(f'<li><a href="{h}">{lbl}</a></li>' for lbl, h in nav_entries) + "</ol></nav>",
    )
    ncx = (
        '<?xml version="1.0" encoding="utf-8"?><ncx xmlns="http://www.daisy.org/z3986/2005/ncx/" version="2005-1"><navMap>'
        + "".join(
            f'<navPoint id="n{i}" playOrder="{i + 1}"><navLabel><text>{lbl}</text></navLabel><content src="{h}"/></navPoint>'
            for i, (lbl, h) in enumerate(nav_entries)
        )
        + "</navMap></ncx>"
    )
    container = book.container or (
        '<?xml version="1.0"?><container version="1.0" xmlns="urn:oasis:names:tc:opendocument:xmlns:container">'
        '<rootfiles><rootfile full-path="OEBPS/content.opf" media-type="application/oebps-package+xml"/></rootfiles></container>'
    )
    with zipfile.ZipFile(path, "w") as z:
        z.writestr("mimetype", "application/epub+zip", compress_type=zipfile.ZIP_STORED)
        z.writestr("META-INF/container.xml", container)
        z.writestr("OEBPS/content.opf", opf)
        if epub3:
            z.writestr("OEBPS/nav.xhtml", nav_xhtml)
        if book.ncx:
            z.writestr("OEBPS/toc.ncx", ncx)
        for d in book.docs:
            z.writestr(f"OEBPS/{d.name}", XHTML.format(title=d.title, body=d.body))
        for name, data in book.media.items():
            z.writestr(f"OEBPS/{name}", data)
        for name, data in book.extra_entries.items():
            z.writestr(zipfile.ZipInfo(name), data)
    return path


@dataclass
class Run:
    rc: int
    out: Path
    stdout: str
    stderr: str

    def chapter(self, prefix: str) -> str:
        files = sorted((self.out / "chapters").glob(f"{prefix}*.md"))
        assert files, f"no chapter {prefix} in {sorted(p.name for p in (self.out / 'chapters').glob('*.md'))}"
        return files[0].read_text(encoding="utf-8")

    def chapters(self) -> list[Path]:
        return sorted((self.out / "chapters").glob("*.md"))

    def all_md(self) -> str:
        return "\n".join(p.read_text(encoding="utf-8") for p in self.chapters())

    @property
    def report(self) -> dict:
        return json.loads((self.out / "conversion-report.json").read_text(encoding="utf-8"))


@pytest.fixture
def convert(tmp_path, capsys):
    counter = {"n": 0}

    def run(book_or_path, *args: str, out: Path | None = None) -> Run:
        counter["n"] += 1
        if isinstance(book_or_path, Book):
            epub = build_epub(tmp_path / f"book{counter['n']}.epub", book_or_path)
        else:
            epub = Path(book_or_path)
        target = out or tmp_path / f"out{counter['n']}"
        rc = main([str(epub), str(target), *args])
        captured = capsys.readouterr()
        return Run(rc, target, captured.out, captured.err)

    return run