Dockerfile

FROM ghcr.io/astral-sh/uv:python3.13-bookworm-slim

ARG GBABELDOCUI_REF=main
ARG BABELDOC_VERSION=0.6.3
ARG PYMUPDF_MIN_VERSION=1.26.7

ENV PDF2ZH_WEB_UI=1 \
    PYTHONUNBUFFERED=1 \
    GRADIO_SERVER_NAME=0.0.0.0 \
    GRADIO_SERVER_PORT=7860 \
    BABELDOC_VERSION=${BABELDOC_VERSION} \
    PYMUPDF_MIN_VERSION=${PYMUPDF_MIN_VERSION}

WORKDIR /app
EXPOSE 7860

RUN apt-get update && \
    apt-get install --no-install-recommends -y \
      ca-certificates \
      curl \
      build-essential \
      libgl1 \
      libglib2.0-0 \
      libxext6 \
      libsm6 \
      libxrender1 && \
    rm -rf /var/lib/apt/lists/*

# Keep the GBabelDocUI frontend, but rebuild it from source so the BabelDOC
# runtime is not trapped inside the stale eaiu/gbabeldocwebui:latest image.
RUN curl -fsSL "https://github.com/eaiu/GBabelDocUI/archive/${GBABELDOCUI_REF}.tar.gz" \
      -o /tmp/gbabeldocui.tar.gz && \
    tar -xzf /tmp/gbabeldocui.tar.gz --strip-components=1 -C /app && \
    rm /tmp/gbabeldocui.tar.gz

# GBabelDocUI still declares the old BabelDOC 0.5 line. Patch only the package
# constraints, then let the normal project install build the same UI entrypoints.
RUN python - <<'PY'
from pathlib import Path
import os

pyproject = Path("pyproject.toml")
text = pyproject.read_text()
replacements = {
    '"pymupdf<1.25.3"': f'"pymupdf>={os.environ["PYMUPDF_MIN_VERSION"]}"',
    '"babeldoc>=0.5.20,<0.6.0"': f'"babeldoc=={os.environ["BABELDOC_VERSION"]}"',
}
for old, new in replacements.items():
    if old not in text:
        raise SystemExit(f"Expected dependency constraint not found: {old}")
    text = text.replace(old, new)
pyproject.write_text(text)
PY

RUN uv pip install --system --no-cache --compile-bytecode .

# Build-time proof: fail the Space build if the deployed kernel is not the
# requested BabelDOC line or if PyMuPDF is still on the old incompatible line.
RUN python - <<'PY'
from importlib.metadata import version, PackageNotFoundError
import os


def parse_version(value: str) -> tuple[int, ...]:
    parts = []
    for token in value.replace('-', '.').split('.'):
        if token.isdigit():
            parts.append(int(token))
        else:
            digits = ''.join(ch for ch in token if ch.isdigit())
            if digits:
                parts.append(int(digits))
            break
    return tuple(parts)

required = os.environ["BABELDOC_VERSION"]
actual = version("babeldoc")
if actual != required:
    raise SystemExit(f"BabelDOC mismatch: expected {required}, got {actual}")

pymupdf_actual = version("pymupdf")
pymupdf_min = os.environ["PYMUPDF_MIN_VERSION"]
if parse_version(pymupdf_actual) < parse_version(pymupdf_min):
    raise SystemExit(f"PyMuPDF too old: expected >= {pymupdf_min}, got {pymupdf_actual}")

for pkg in ["pdf2zh-gbabeldocui", "babeldoc", "pymupdf", "cryptography", "onnx"]:
    try:
        print(f"{pkg}=={version(pkg)}")
    except PackageNotFoundError:
        print(f"{pkg}=not-installed")
PY

RUN babeldoc --version && babeldoc --warmup && pdf2zh --version

CMD ["pdf2zh", "--gui"]