50 lines
1.3 KiB
Text
50 lines
1.3 KiB
Text
FROM python:3.13-slim
|
||
|
||
# Install curl for uv installation and openssl for RSA key generation
|
||
# Also install git for potential dependencies and build-essential for native compilations
|
||
RUN apt-get update && apt-get install -y \
|
||
curl \
|
||
openssl \
|
||
git \
|
||
build-essential \
|
||
&& rm -rf /var/lib/apt/lists/*
|
||
|
||
# Install uv
|
||
RUN curl -LsSf https://astral.sh/uv/install.sh | sh
|
||
ENV PATH="/root/.local/bin:$PATH"
|
||
|
||
# Set working directory
|
||
WORKDIR /app
|
||
|
||
# Copy Python dependencies
|
||
COPY pyproject.toml uv.lock ./
|
||
RUN uv sync
|
||
|
||
# Copy sample document and warmup script for docling
|
||
COPY documents/warmup_ocr.pdf ./
|
||
COPY warm_up_docling.py ./
|
||
RUN uv run docling-tools models download
|
||
RUN uv run python - <<'PY'
|
||
import pathlib, easyocr
|
||
cache = pathlib.Path("/root/.EasyOCR/model")
|
||
cache.mkdir(parents=True, exist_ok=True)
|
||
# Prewarm the detector + recog for Docling’s default langs
|
||
easyocr.Reader(['fr','de','es','en'],
|
||
download_enabled=True,
|
||
model_storage_directory=str(cache))
|
||
print("EasyOCR cache ready at", cache)
|
||
PY
|
||
|
||
# RUN uv run python warm_up_docling.py && rm warm_up_docling.py warmup_ocr.pdf
|
||
|
||
|
||
#ENV EASYOCR_MODULE_PATH=~/.cache/docling/models/EasyOcr/
|
||
|
||
# Copy Python source
|
||
COPY src/ ./src/
|
||
|
||
# Expose backend port
|
||
EXPOSE 8000
|
||
|
||
# Start backend in foreground
|
||
CMD ["uv", "run", "python", "src/main.py"]
|