Dockerfile.dev

# syntax=docker/dockerfile:1.4
FROM python:3.11-slim-trixie

WORKDIR /app

ARG TARGETARCH
ARG CHROMIUM_VERSION=149.0.7827.196-1~deb13u1
ARG CHROMIUM_SNAPSHOT=20260625T180000Z

# LiteParse (Node + @llamaindex/liteparse) for document extraction; OCR via Tesseract.
ENV APP_DATA_DIRECTORY=/app_data \
    TEMP_DIRECTORY=/tmp/presenton \
    UV_SYSTEM_PYTHON=1 \
    UV_COMPILE_BYTECODE=1 \
    UV_LINK_MODE=copy \
    PATH="/root/.local/bin:${PATH}" \
    EXPORT_PACKAGE_ROOT=/app/presentation-export \
    EXPORT_RUNTIME_DIR=/app/presentation-export \
    BUILT_PYTHON_MODULE_PATH=/app/presentation-export/py/convert-linux-current \
    PRESENTON_APP_ROOT=/app \
    HF_HOME=/root/.cache/huggingface \
    PRESENTON_FASTEMBED_ICON_CACHE_DIR=/root/.cache/presenton/fastembed-icons \
    PUPPETEER_EXECUTABLE_PATH=/usr/bin/chromium

RUN printf 'Acquire::Check-Valid-Until "false";\n' > /etc/apt/apt.conf.d/99snapshot \
    && printf 'deb [check-valid-until=no] http://snapshot.debian.org/archive/debian-security/%s trixie-security main\n' "$CHROMIUM_SNAPSHOT" > /etc/apt/sources.list.d/chromium-snapshot.list \
    && apt-get update && apt-get install -y --no-install-recommends --allow-downgrades \
    ca-certificates curl unzip \
    nginx fontconfig imagemagick zstd \
    fonts-liberation fonts-noto-core fonts-noto-extra fonts-noto-mono fonts-noto-ui-core fonts-noto-ui-extra \
    fonts-noto-cjk fonts-noto-cjk-extra fonts-noto-color-emoji xdg-utils \
    libasound2t64 libatk-bridge2.0-0t64 libatk1.0-0t64 libatspi2.0-0t64 \
    libcairo2 libcups2t64 libdbus-1-3 libdrm2 libexpat1 libgbm1 \
    libglib2.0-0t64 libgtk-3-0t64 libnspr4 libnss3 libpango-1.0-0 \
    libx11-6 libxcb1 libxcomposite1 libxdamage1 libxext6 libxfixes3 \
    libxkbcommon0 libxrandr2 libxshmfence1 libxss1 libxtst6 \
    tesseract-ocr tesseract-ocr-eng \
    chromium="${CHROMIUM_VERSION}" \
    chromium-common="${CHROMIUM_VERSION}" \
    chromium-driver="${CHROMIUM_VERSION}" \
    && apt-mark hold chromium chromium-common chromium-driver \
    && curl -LsSf https://astral.sh/uv/install.sh | sh \
    && rm -rf /var/lib/apt/lists/*

# Remove any non-Noto fonts that may have been installed as dependencies.
RUN find /usr/share/fonts -type f ! -iname 'Noto*' -delete \
    && find /usr/share/fonts -type d -empty -delete \
    && fc-cache -fsv

RUN curl -fsSL https://deb.nodesource.com/setup_20.x | bash - \
    && apt-get install -y nodejs \
    && rm -rf /var/lib/apt/lists/*

COPY package.json package-lock.json /app/
RUN npm --prefix /app install --omit=dev

RUN mkdir -p /app/document-extraction-liteparse \
    && cd /app/document-extraction-liteparse \
    && npm init -y \
    && npm install @llamaindex/liteparse@1.5.2 --omit=dev
COPY electron/resources/document-extraction/liteparse_runner.mjs /app/document-extraction-liteparse/liteparse_runner.mjs

COPY scripts/sync-presentation-export.cjs /app/scripts/sync-presentation-export.cjs
COPY scripts/presenton-terminal-banner.mjs /app/scripts/presenton-terminal-banner.mjs
COPY scripts/user-config-env.cjs /app/scripts/user-config-env.cjs
RUN rm -rf /app/presentation-export \
    && EXPORT_RUNTIME_ARCH="${TARGETARCH}" node /app/scripts/sync-presentation-export.cjs --force \
    && find /app/presentation-export/py -maxdepth 1 -type f -name "convert-linux-*" -exec chmod +x {} \; \
    && set -eux; \
    if [ -z "${TARGETARCH:-}" ]; then TARGETARCH="$(dpkg --print-architecture)"; fi; \
    case "$TARGETARCH" in \
    amd64) export_arch="x64" ;; \
    arm64) export_arch="arm64" ;; \
    *) echo "Unsupported TARGETARCH: $TARGETARCH" && exit 1 ;; \
    esac; \
    test -f "/app/presentation-export/py/convert-linux-${export_arch}"; \
    ln -sf "/app/presentation-export/py/convert-linux-${export_arch}" /app/presentation-export/py/convert-linux-current; \
    chmod +x "/app/presentation-export/py/convert-linux-${export_arch}"; \
    ls -lah /app/presentation-export/py

# Bind mount `.:/app` hides any .venv under servers/fastapi at runtime — install deps into
# system site-packages (same interpreter `start.js` uses as `python`).
COPY servers/fastapi /app/servers/fastapi
COPY templates /app/templates
WORKDIR /app/servers/fastapi
RUN --mount=type=cache,target=/root/.cache/uv \
    uv export --frozen --no-dev --no-emit-project -o /tmp/requirements.txt \
    && uv pip install --system -r /tmp/requirements.txt \
    && uv pip install --system --no-deps .
# Mem0 BM25 lemmatization requires a spaCy language model at runtime.
# Installing it in the image avoids per-run downloads and startup self-disable.
RUN --mount=type=cache,target=/root/.cache/uv \
    uv pip install --system \
    "https://github.com/explosion/spacy-models/releases/download/en_core_web_sm-3.8.0/en_core_web_sm-3.8.0-py3-none-any.whl"

# FastEmbed + Hugging Face Hub weights for icon search and Mem0 (no cache-only mount here:
# those files must remain in the image layer).
WORKDIR /app/servers/fastapi
RUN python scripts/warm_fastembed_cache.py

WORKDIR /app
COPY nginx.conf /etc/nginx/nginx.conf

EXPOSE 80
CMD ["node", "/app/start.js", "--dev"]