Dockerfile

# 使用 OpenJDK 17 作为基础镜像
FROM openjdk:17-jdk-slim AS builder

# 设置工作目录
WORKDIR /app

# 安装必要的构建工具
RUN apt-get update && apt-get install -y \
    git \
    curl \
    gradle \
    maven

# 克隆项目代码
RUN git clone https://github.com/Stirling-Tools/Stirling-PDF.git .

# 构建项目
RUN ./gradlew build

# 运行阶段
FROM openjdk:17-jdk-slim

# 创建应用用户和必要目录
RUN useradd -m -d /home/appuser appuser && \
    mkdir -p /home/appuser/.cache/dconf && \
    chown -R appuser:appuser /home/appuser && \
    mkdir -p /app/logs /app/configs /app/customFiles /usr/share/tessdata && \
    chmod -R 777 /app/logs /app/configs /app/customFiles /usr/share/tessdata

# 设置工作目录
WORKDIR /app

# 复制构建的 JAR 文件到运行镜像
COPY --from=builder /app/build/libs/*.jar /app/stirling-pdf.jar

# 安装运行时依赖
RUN apt-get update && apt-get install -y \
    libreoffice \
    poppler-utils \
    tesseract-ocr \
    tesseract-ocr-eng \
    wget \
    python3 \
    python3-pip && \
    pip3 install --no-cache-dir unoconv WeasyPrint pdf2image pillow && \
    cd /usr/share/tessdata && \
    wget https://github.com/tesseract-ocr/tessdata/raw/main/chi_sim.traineddata && \
    wget https://github.com/tesseract-ocr/tessdata/raw/main/chi_tra.traineddata && \
    wget https://github.com/tesseract-ocr/tessdata/raw/main/eng.traineddata && \
    wget https://github.com/tesseract-ocr/tessdata/raw/main/jpn.traineddata && \
    wget https://github.com/tesseract-ocr/tessdata/raw/main/kor.traineddata && \
    wget https://github.com/tesseract-ocr/tessdata/raw/main/rus.traineddata && \
    wget https://github.com/tesseract-ocr/tessdata/raw/main/fra.traineddata && \
    wget https://github.com/tesseract-ocr/tessdata/raw/main/deu.traineddata && \
    wget https://github.com/tesseract-ocr/tessdata/raw/main/spa.traineddata && \
    wget https://github.com/tesseract-ocr/tessdata/raw/main/ita.traineddata && \
    wget https://github.com/tesseract-ocr/tessdata/raw/main/por.traineddata && \
    wget https://github.com/tesseract-ocr/tessdata/raw/main/vie.traineddata && \
    wget https://github.com/tesseract-ocr/tessdata/raw/main/tha.traineddata && \
    rm -rf /var/lib/apt/lists/*

# 设置环境变量
ENV HOME=/home/appuser \
    XDG_RUNTIME_DIR=/tmp/runtime-appuser \
    LIBREOFFICE_CONFIG_DIR=/tmp/libreoffice

# 设置目录权限
RUN mkdir -p ${XDG_RUNTIME_DIR} ${LIBREOFFICE_CONFIG_DIR} && \
    chown -R appuser:appuser ${XDG_RUNTIME_DIR} ${LIBREOFFICE_CONFIG_DIR} && \
    chmod 777 ${XDG_RUNTIME_DIR} ${LIBREOFFICE_CONFIG_DIR}

# 切换用户
USER appuser

# 暴露端口 7860
EXPOSE 7860

# 启动应用程序
CMD ["java", "-Dserver.port=7860", "-Dserver.address=0.0.0.0", "-Dfile.encoding=UTF-8", "-jar", "/app/stirling-pdf.jar"]