Dockerfile

FROM python:3.12-slim@sha256:c3d81d25b3154142b0b42eb1e61300024426268edeb5b5a26dd7ddf64d9daf28

ENV DEBIAN_FRONTEND=noninteractive

RUN apt-get update \
    && apt-get install -y --no-install-recommends \
        bash \
        ca-certificates \
        coreutils \
        curl \
        file \
        findutils \
        jq \
        less \
        poppler-utils \
        ripgrep \
        unzip \
        zip \
    && rm -rf /var/lib/apt/lists/*

RUN pip install --no-cache-dir \
    openpyxl==3.1.5 \
    pandas==3.0.3 \
    pdfplumber==0.11.10 \
    pypdf==6.14.2 \
    pytest==8.3.4 \
    python-docx==1.2.0 \
    python-pptx==1.0.2 \
    pyyaml==6.0.2 \
    reportlab==5.0.0

ENV NVM_DIR=/root/.nvm
ENV NODE_VERSION=22.23.1

SHELL ["/bin/bash", "-o", "pipefail", "-c"]

RUN curl -fsSL --retry 5 --retry-delay 2 \
        https://raw.githubusercontent.com/nvm-sh/nvm/v0.40.2/install.sh \
        | bash \
    && source "$NVM_DIR/nvm.sh" \
    && nvm install "$NODE_VERSION" \
    && nvm alias default "$NODE_VERSION" \
    && nvm use "$NODE_VERSION" \
    && npm install -g openclaw@2026.7.1 \
        --fetch-retries=5 \
        --fetch-retry-mintimeout=20000 \
        --fetch-retry-maxtimeout=120000 \
    && npm cache clean --force

ENV PATH=/root/.nvm/versions/node/v22.23.1/bin:${PATH}

RUN source "$NVM_DIR/nvm.sh" \
    && nvm use 22 \
    && test "$(node --version)" = "v22.23.1" \
    && npm list -g openclaw@2026.7.1 --depth=0 \
    && openclaw --version | grep -Eq '(^| )2026\.7\.1($| )'

RUN printf '%s\n' '#!/bin/sh' 'exec python /app/tools/xlsx_dump.py "$@"' \
        > /usr/local/bin/xlsx-dump \
    && chmod +x /usr/local/bin/xlsx-dump \
    && ln -s /app/pdf_text_extract.py /usr/local/bin/pdf_text_extract.py \
    && ln -s /app/baseline_manifest.txt /opt/baseline_manifest.txt \
    && mkdir -p /app/data /app/output

WORKDIR /app

CMD ["python", "-m", "http.server", "7860", "--bind", "0.0.0.0"]