Dockerfile.worker

# SPDX-FileCopyrightText: 2026 ArcheBase
#
# SPDX-License-Identifier: MulanPSL-2.0

# Multi-stage Dockerfile for building the Roboflow worker container.
#
# This Dockerfile creates a minimal container image for running Roboflow
# data conversion workers in Kubernetes environments.
#
# Build:
#   docker build -f Dockerfile.worker -t roboflow-worker:latest .
#
# Run:
#   docker run --rm roboflow-worker:latest convert --help

# =============================================================================
# Builder stage: Compile the Rust binary with all features
# =============================================================================
ARG RUST_VERSION=1.83
FROM docker.io/library/rust:${RUST_VERSION}-slim AS builder

# Install build dependencies
RUN apt-get update && apt-get install -y \
    pkg-config \
    libssl-dev \
    ca-certificates \
    curl \
    && rm -rf /var/lib/apt/lists/*

# Set working directory
WORKDIR /build

# Copy Cargo manifests and lock file
COPY Cargo.toml Cargo.lock ./
COPY crates/ ./crates/
COPY src/ ./src/

# Build the worker binary with release optimizations and all features
# - dataset-all: Enable all dataset formats (HDF5, Parquet, depth)
# - distributed: Enable TiKV distributed coordination
# - cloud-storage: Enable S3/OSS cloud storage support
# - jemalloc: Use jemalloc for better multi-threaded performance
RUN --mount=type=cache,target=/usr/local/cargo/registry \
    --mount=type=cache,target=/target \
    cargo build --release \
    --bins \
    --features "dataset-all,distributed,cloud-storage,jemalloc" \
    && mkdir -p /out \
    && cp target/release/convert /out/ \
    && cp target/release/extract /out/ \
    && cp target/release/inspect /out/ \
    && cp target/release/schema /out/ \
    && cp target/release/search /out/

# =============================================================================
# Runtime stage: Minimal image with only runtime dependencies
# =============================================================================
FROM docker.io/library/debian:bookworm-slim AS runtime

# Install runtime dependencies:
# - ffmpeg: For video encoding/decoding
# - ca-certificates: For HTTPS connections
# - libhdf5-*, libsz2, libzstd1, liblz4-1, libb2t1: HDF5 and compression libraries
# - libssl3, libcrypto++3: SSL/TLS support
RUN apt-get update && apt-get install -y \
    ffmpeg \
    ca-certificates \
    libhdf5-103 \
    libhdf5-cpp-103 \
    libsz2 \
    libzstd1 \
    liblz4-1 \
    libbz2-1 \
    libssl3 \
    libcurl4 \
    libgomp1 \
    && rm -rf /var/lib/apt/lists/* \
    && update-ca-certificates

# Create non-root user for running the worker
RUN groupadd -r roboflow && useradd -r -g roboflow -u 1001 roboflow

# Set working directory
WORKDIR /workspace

# Copy binaries from builder
COPY --from=builder --chown=roboflow:roboflow /out/ /usr/local/bin/

# Verify the binary was copied
RUN chmod +x /usr/local/bin/convert \
    /usr/local/bin/extract \
    /usr/local/bin/inspect \
    /usr/local/bin/schema \
    /usr/local/bin/search

# Switch to non-root user
USER roboflow

# Set environment variables
ENV RUST_LOG=info \
    RUST_BACKTRACE=1 \
    PYTHONUNBUFFERED=1

# Health check to verify the container is functional
HEALTHCHECK --interval=30s --timeout=10s --start-period=5s --retries=3 \
    CMD convert --version || exit 1

# Default command shows help
ENTRYPOINT ["/usr/local/bin/convert"]
CMD ["--help"]