# syntax=docker/dockerfile:1
# Stage 1: Build Vue frontend
FROM node:22-alpine AS build-frontend
WORKDIR /build
COPY frontend/package.json frontend/package-lock.json* ./
RUN npm ci --quiet
COPY frontend/ .
RUN npm run build

# Stage 2: Python runtime
# Tracks CI image (ci-python:3.14) so test results stay representative.
FROM python:3.14-slim AS runtime
WORKDIR /app

COPY pyproject.toml .
COPY src/ src/
RUN --mount=type=cache,target=/root/.cache/pip \
    pip install .

# Speech-to-text (faster-whisper + soundfile). faster-whisper is pure
# Python; the actual inference engine is ctranslate2 (C++) which has
# cp314 wheels as of v4.7.2 (2026-05-19). No torch needed — ctranslate2
# does its own CPU inference. Image add: ~150 MB.
RUN --mount=type=cache,target=/root/.cache/pip \
    pip install faster-whisper soundfile

# Text-to-speech (piper-tts). Replaces kokoro, which has been
# stale upstream since April 2025 (requires_python<3.13). Piper depends
# only on onnxruntime (already pulled in for STT via faster-whisper) and
# pathvalidate — total Python overhead is tiny. Voice models are separate
# .onnx + .onnx.json files bundled below.
RUN --mount=type=cache,target=/root/.cache/pip \
    pip install piper-tts

# Bundle two default voices in the image so first-run TTS works offline.
# Additional voices can be downloaded at runtime into /data/voices via the
# admin UI (see services/tts.py for the voice-discovery logic).
# Voice catalog: https://huggingface.co/rhasspy/piper-voices
#
# Using Python's urllib instead of curl/wget because python:3.14-slim
# ships neither; python is obviously available. The heredoc requires
# the BuildKit Dockerfile 1.3+ frontend, which the `# syntax=...:1`
# directive at the top of this file already pulls in.
RUN <<'PYEOF' python3
import os, urllib.request

VOICES = ["en_US-amy-medium", "en_US-ryan-medium"]
BASE = "https://huggingface.co/rhasspy/piper-voices/resolve/main/en/en_US"
TARGET = "/opt/piper-voices"
os.makedirs(TARGET, exist_ok=True)
for v in VOICES:
    dataset = v.split("-")[1]
    for ext in ("onnx", "onnx.json"):
        url = f"{BASE}/{dataset}/medium/{v}.{ext}"
        path = os.path.join(TARGET, f"{v}.{ext}")
        print(f"Downloading {url}", flush=True)
        urllib.request.urlretrieve(url, path)
        size = os.path.getsize(path)
        print(f"  -> {path} ({size:,} bytes)", flush=True)
PYEOF

# Build the fable-mcp wheel so it can be served for download
COPY fable-mcp/ fable-mcp/
RUN --mount=type=cache,target=/root/.cache/pip \
    pip install build hatchling \
    && python -m build --wheel ./fable-mcp --outdir /app/dist/ \
    && pip uninstall -y build \
    && rm -rf fable-mcp/

COPY --from=build-frontend /build/dist/ src/fabledassistant/static/
COPY alembic.ini .
COPY alembic/ alembic/

# Ensure Python finds the source tree (where static files live) before site-packages
ENV PYTHONPATH=/app/src

# Version is injected at build time via --build-arg BUILD_VERSION=YY.MM.DD.N
# Falls back to "dev" for local / untagged builds
ARG BUILD_VERSION=dev
ENV APP_VERSION=$BUILD_VERSION

EXPOSE 5000
CMD ["sh", "-c", "alembic upgrade head && hypercorn 'fabledassistant.app:create_app()' --bind 0.0.0.0:5000 --keep-alive 600"]