External/SnapOtter
1
0
Fork 0
mirror of https://github.com/snapotter-hq/SnapOtter.git synced 2026-08-03 07:46:42 +02:00
Code Issues Packages Projects Releases Wiki Activity

feat: lightweight Docker image without AI/ML tools (:lite tag)

Browse Source 
Closes #1
This commit is contained in:
stirling-image
2026-04-05 00:23:21 +08:00
committed by GitHub
parent 51f10abb35
commit 449a2fc319
19 changed files with 1819 additions and 72 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docker/Dockerfile
+37 -27
View File
@@ -4,6 +4,8 @@
# Multi-stage build for single-container deployment
# ============================================
ARG VARIANT=full
# ============================================
# Stage 1: Build the frontend (Vite + React)
# ============================================
@@ -40,52 +42,56 @@ RUN --mount=type=cache,id=turbo-cache,target=/app/.turbo \
# ============================================
FROM node:22-bookworm AS production
ARG VARIANT
RUN corepack enable && corepack prepare pnpm@9.15.4 --activate
# Install system dependencies for image processing and AI
# System dependencies shared by all variants
RUN apt-get update && apt-get install -y --no-install-recommends \
python3 python3-pip python3-venv python3-dev \
imagemagick \
tesseract-ocr tesseract-ocr-eng tesseract-ocr-deu tesseract-ocr-fra tesseract-ocr-spa \
libraw-dev \
potrace \
curl \
build-essential \
libgl1 libglib2.0-0 \
gosu \
libheif-examples \
&& rm -rf /var/lib/apt/lists/*
# Create Python venv and install ML packages
RUN python3 -m venv /opt/venv
# Python/ML system dependencies (full variant only)
RUN if [ "$VARIANT" = "full" ]; then \
apt-get update && apt-get install -y --no-install-recommends \
python3 python3-pip python3-venv python3-dev \
tesseract-ocr tesseract-ocr-eng tesseract-ocr-deu tesseract-ocr-fra tesseract-ocr-spa \
build-essential \
libgl1 libglib2.0-0 \
&& rm -rf /var/lib/apt/lists/* \
; fi
# Python venv + ML packages + model weights (full variant only)
COPY packages/ai/python/requirements.txt /tmp/requirements.txt
# Install Python packages - fail loudly for critical ones, warn for optional
RUN --mount=type=cache,id=pip-cache,target=/root/.cache/pip \
RUN if [ "$VARIANT" = "full" ]; then \
python3 -m venv /opt/venv && \
/opt/venv/bin/pip install --upgrade pip && \
/opt/venv/bin/pip install \
Pillow numpy opencv-python-headless onnxruntime && \
(/opt/venv/bin/pip install rembg[cpu] || echo "WARNING: rembg not installed - background removal will be unavailable") && \
(/opt/venv/bin/pip install realesrgan || echo "WARNING: realesrgan not installed - will fallback to Lanczos upscaling") && \
(/opt/venv/bin/pip install paddlepaddle paddleocr || echo "WARNING: PaddleOCR not installed - will fallback to Tesseract") && \
(/opt/venv/bin/pip install mediapipe || echo "WARNING: mediapipe not installed - face detection will be unavailable") && \
(/opt/venv/bin/pip install lama-cleaner || echo "WARNING: lama-cleaner not installed - object eraser will be unavailable") && \
rm /tmp/requirements.txt
Pillow numpy opencv-python-headless onnxruntime && \
(/opt/venv/bin/pip install "rembg[cpu]" || echo "WARNING: rembg not installed") && \
(/opt/venv/bin/pip install realesrgan || echo "WARNING: realesrgan not installed") && \
(/opt/venv/bin/pip install paddlepaddle paddleocr || echo "WARNING: PaddleOCR not installed") && \
(/opt/venv/bin/pip install mediapipe || echo "WARNING: mediapipe not installed") && \
(/opt/venv/bin/pip install lama-cleaner || echo "WARNING: lama-cleaner not installed") \
; fi && rm -f /tmp/requirements.txt
# Pre-download ALL AI model weights into the image (no first-use download delays)
# This makes the Docker image fully self-contained — works offline
COPY docker/download_models.py /tmp/download_models.py
RUN /opt/venv/bin/python3 /tmp/download_models.py && rm /tmp/download_models.py
RUN /opt/venv/bin/python3 -c "\
RUN if [ "$VARIANT" = "full" ]; then \
/opt/venv/bin/python3 /tmp/download_models.py && \
/opt/venv/bin/python3 -c "\
try: \
from paddleocr import PaddleOCR; \
print('Downloading PaddleOCR models...'); \
ocr = PaddleOCR(use_angle_cls=True, lang='en', show_log=False); \
print('PaddleOCR models ready'); \
except: print('PaddleOCR model pre-download skipped') \
" 2>/dev/null || echo "WARNING: Could not pre-download PaddleOCR models"
" 2>/dev/null || echo "WARNING: Could not pre-download PaddleOCR models" \
; fi && rm -f /tmp/download_models.py
WORKDIR /app
@@ -105,8 +111,10 @@ RUN --mount=type=cache,id=pnpm-store,target=/root/.local/share/pnpm/store/v3 \
pnpm install --frozen-lockfile --prod
# Remove build tools no longer needed in production
RUN apt-get purge -y --auto-remove build-essential python3-dev && \
rm -rf /var/lib/apt/lists/*
RUN if [ "$VARIANT" = "full" ]; then \
apt-get purge -y --auto-remove build-essential python3-dev && \
rm -rf /var/lib/apt/lists/* \
; fi
# Copy source code for API (tsx runs TS directly - no build step needed)
COPY apps/api/src ./apps/api/src
@@ -143,11 +151,13 @@ ENV PORT=1349 \
MAX_BATCH_SIZE=200 \
CONCURRENT_JOBS=3 \
MAX_MEGAPIXELS=100 \
RATE_LIMIT_PER_MIN=100
RATE_LIMIT_PER_MIN=100 \
STIRLING_VARIANT=${VARIANT}
# Create non-root user for runtime
RUN groupadd -r stirling && useradd -r -g stirling -d /app -s /sbin/nologin stirling
RUN chown -R stirling:stirling /app /data /tmp/workspace /opt/venv
RUN chown -R stirling:stirling /app /data /tmp/workspace && \
([ -d /opt/venv ] && chown -R stirling:stirling /opt/venv || true)
# Entrypoint fixes volume permissions then drops to stirling via gosu
COPY docker/entrypoint.sh /usr/local/bin/entrypoint.sh