Dockerfile

# syntax=docker/dockerfile:1.7
# ---------------------------------------------------------------------------
# Production worker image — multi-stage build.
# Keeps the OCR system packages (tesseract, poppler, libgl1) in runtime;
# drops build-essential and the [dev] python extras.
# ---------------------------------------------------------------------------

FROM python:3.12-slim AS builder

ENV PYTHONUNBUFFERED=1 \
    PYTHONDONTWRITEBYTECODE=1 \
    PIP_NO_CACHE_DIR=1 \
    PIP_DISABLE_PIP_VERSION_CHECK=1

WORKDIR /build

RUN apt-get update && apt-get install -y --no-install-recommends \
        build-essential \
    && rm -rf /var/lib/apt/lists/*

RUN python -m venv /opt/venv
ENV PATH="/opt/venv/bin:$PATH"

COPY pyproject.toml README.md ./
RUN mkdir -p app && touch app/__init__.py \
    && pip install --upgrade pip wheel setuptools \
    && pip install . \
    && (pip install ".[worker]" || \
       (echo "WARN: worker extras failed; installing fallbacks" \
        && pip install pytesseract pdf2image Pillow numpy))


# ---------------------------------------------------------------------------
FROM python:3.12-slim AS runtime

ENV PYTHONUNBUFFERED=1 \
    PYTHONDONTWRITEBYTECODE=1 \
    PYTHONPATH=/app \
    PATH="/opt/venv/bin:$PATH" \
    HF_HOME=/app/.cache/huggingface \
    XDG_CACHE_HOME=/app/.cache

# Runtime-only OCR/video system packages (no build-essential).
RUN apt-get update && apt-get install -y --no-install-recommends \
        curl \
        tesseract-ocr \
        tesseract-ocr-eng \
        poppler-utils \
        libgl1 \
        libglib2.0-0 \
        ffmpeg \
    && rm -rf /var/lib/apt/lists/* \
    && addgroup --system --gid 1001 app \
    && adduser --system --uid 1001 --ingroup app --no-create-home app

WORKDIR /app

# Self-contained venv from the builder stage.
COPY --from=builder /opt/venv /opt/venv

# Pre-download the three retrieval models (embedder, reranker, BM25 sparse).
# Retry on HF 429s; FAIL the build if retries exhaust so the CI error is
# loud instead of producing an image whose worker can't embed/rerank.
RUN mkdir -p /app/.cache/huggingface && \
    for i in 1 2 3 4 5; do \
      if python -c "\
from sentence_transformers import SentenceTransformer, CrossEncoder; \
from fastembed import SparseTextEmbedding; \
SentenceTransformer('BAAI/bge-small-en-v1.5'); \
CrossEncoder('BAAI/bge-reranker-base'); \
list(SparseTextEmbedding(model_name='Qdrant/bm25').embed(['warmup']))\
"; then \
        break; \
      fi; \
      if [ "$i" = "5" ]; then \
        echo "FATAL: HF model download failed after 5 attempts"; exit 1; \
      fi; \
      echo "HF download attempt $i failed; sleeping $((i*15))s"; \
      sleep $((i*15)); \
    done

COPY app ./app
COPY alembic ./alembic
COPY alembic.ini ./alembic.ini
RUN chown -R app:app /app

USER app

EXPOSE 7860

CMD ["python", "-m", "app.workers.main"]