[Rust] Standalone vector index build and ANN search with Python API #23
Workflow file for this run
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| # SPDX-FileCopyrightText: 2026 LakeSoul Contributors | |
| # | |
| # SPDX-License-Identifier: Apache-2.0 | |
| name: Compatibility CI | |
| on: | |
| pull_request: | |
| paths: | |
| - ".github/workflows/compatibility-ci.yml" | |
| - "Cargo.toml" | |
| - "rust/**" | |
| - "python/**" | |
| - "pom.xml" | |
| - "lakesoul-common/**" | |
| - "lakesoul-spark/**" | |
| - "lakesoul-flink/**" | |
| - "native-io/**" | |
| branches: | |
| - "main" | |
| - "release/**" | |
| workflow_dispatch: | |
| inputs: | |
| storage: | |
| description: "Storage backend to validate" | |
| required: true | |
| default: "file" | |
| type: choice | |
| options: | |
| - "file" | |
| - "rustfs" | |
| - "all" | |
| engines: | |
| description: "Engine group to validate" | |
| required: true | |
| default: "all" | |
| type: choice | |
| options: | |
| - "all" | |
| - "python" | |
| - "jvm" | |
| - "rust" | |
| cases: | |
| description: "Case breadth" | |
| required: true | |
| default: "smoke" | |
| type: choice | |
| options: | |
| - "smoke" | |
| - "full" | |
| env: | |
| RUSTFLAGS: "-Awarnings" | |
| LAKESOUL_PG_URL: "jdbc:postgresql://127.0.0.1:5432/lakesoul_test?stringtype=unspecified" | |
| LAKESOUL_PG_USERNAME: "lakesoul_test" | |
| LAKESOUL_PG_PASSWORD: "lakesoul_test" | |
| LAKESOUL_IO_USE_V2_MERGE: "true" | |
| jobs: | |
| compatibility-file: | |
| name: File Storage Compatibility | |
| runs-on: ubuntu-24.04 | |
| if: ${{ github.event_name != 'workflow_dispatch' || inputs.storage == 'file' || inputs.storage == 'all' }} | |
| services: | |
| postgres: | |
| image: postgres:14.5 | |
| env: | |
| POSTGRES_PASSWORD: lakesoul_test | |
| POSTGRES_USER: lakesoul_test | |
| POSTGRES_DB: lakesoul_test | |
| options: >- | |
| --health-cmd pg_isready | |
| --health-interval 10s | |
| --health-timeout 5s | |
| --health-retries 5 | |
| --name lakesoul-compat-pg | |
| ports: | |
| - 5432:5432 | |
| steps: | |
| - name: Free Disk Space (Ubuntu) | |
| uses: jlumbroso/free-disk-space@main | |
| with: | |
| tool-cache: false | |
| android: true | |
| dotnet: true | |
| haskell: true | |
| large-packages: false | |
| docker-images: true | |
| swap-storage: true | |
| - uses: actions/checkout@v4 | |
| - name: Set up JDK 11 | |
| uses: actions/setup-java@v4 | |
| with: | |
| java-version: "11" | |
| distribution: "temurin" | |
| cache: maven | |
| - name: Set up Python 3.10 | |
| uses: actions/setup-python@v5 | |
| with: | |
| python-version: "3.10" | |
| - name: Install Rust toolchain | |
| uses: dtolnay/rust-toolchain@master | |
| with: | |
| toolchain: stable | |
| - name: Install Protoc | |
| uses: arduino/setup-protoc@v2 | |
| with: | |
| version: "23.x" | |
| repo-token: ${{ secrets.GITHUB_TOKEN }} | |
| - name: Install uv | |
| uses: astral-sh/setup-uv@v6 | |
| with: | |
| enable-cache: true | |
| cache-suffix: "lakesoul-compat-ci" | |
| cache-dependency-glob: | | |
| python/pyproject.toml | |
| python/uv.lock | |
| - uses: Swatinem/rust-cache@v2 | |
| with: | |
| workspaces: ". -> rust/target" | |
| env-vars: "JAVA_HOME" | |
| - name: Install psql | |
| run: sudo apt-get install -y postgresql-client-16 | |
| - name: Init PostgreSQL metadata | |
| run: ./script/meta_init_for_local_test.sh -j 1 | |
| - name: Build native libraries | |
| run: cargo -q build --release -p lakesoul-io-c -p lakesoul-metadata-c | |
| - name: Build Spark and Flink test artifacts | |
| run: mvn -q -B package -pl lakesoul-spark,lakesoul-flink -am -Pcross-build -DskipTests --file pom.xml | |
| - name: Build Python package | |
| working-directory: python | |
| run: | | |
| uv sync --python 3.10 --extra pandas --extra ray --extra daft --extra pyspark | |
| - name: Run compatibility matrix | |
| env: | |
| LAKESOUL_SOURCE_DIR: ${{ github.workspace }} | |
| PYTHONPATH: tests:src | |
| working-directory: python | |
| run: | | |
| EVENT_NAME="${{ github.event_name }}" | |
| if [ "$EVENT_NAME" = "workflow_dispatch" ]; then | |
| MODE="${{ inputs.cases }}" | |
| ENGINES="${{ inputs.engines }}" | |
| elif [ "$EVENT_NAME" = "schedule" ]; then | |
| MODE="full" | |
| ENGINES="all" | |
| else | |
| MODE="smoke" | |
| ENGINES="all" | |
| fi | |
| .venv/bin/python -m compat.run_matrix \ | |
| --mode "$MODE" \ | |
| --cases "$MODE" \ | |
| --engines "$ENGINES" \ | |
| --storage "file:///tmp/lakesoul-compat" \ | |
| --output-dir "${{ github.workspace }}/compatibility-artifacts/file" | |
| - name: Dump PostgreSQL metadata | |
| if: always() | |
| run: | | |
| mkdir -p compatibility-artifacts/file/postgres | |
| PGPASSWORD=lakesoul_test psql -h 127.0.0.1 -p 5432 -U lakesoul_test -d lakesoul_test -c "select * from table_info;" > compatibility-artifacts/file/postgres/table_info.txt | |
| PGPASSWORD=lakesoul_test psql -h 127.0.0.1 -p 5432 -U lakesoul_test -d lakesoul_test -c "select * from data_commit_info;" > compatibility-artifacts/file/postgres/data_commit_info.txt | |
| - name: Upload compatibility artifacts | |
| if: always() | |
| uses: actions/upload-artifact@v4 | |
| with: | |
| name: compatibility-file-artifacts | |
| path: compatibility-artifacts/file | |
| retention-days: 7 | |
| if-no-files-found: warn | |
| compatibility-rustfs: | |
| name: RustFS Compatibility | |
| runs-on: ubuntu-24.04 | |
| if: ${{ github.event_name == 'schedule' || (github.event_name == 'workflow_dispatch' && (inputs.storage == 'rustfs' || inputs.storage == 'all')) }} | |
| services: | |
| postgres: | |
| image: postgres:14.5 | |
| env: | |
| POSTGRES_PASSWORD: lakesoul_test | |
| POSTGRES_USER: lakesoul_test | |
| POSTGRES_DB: lakesoul_test | |
| options: >- | |
| --health-cmd pg_isready | |
| --health-interval 10s | |
| --health-timeout 5s | |
| --health-retries 5 | |
| --name lakesoul-compat-rustfs-pg | |
| ports: | |
| - 5432:5432 | |
| rustfs: | |
| image: swr.cn-southwest-2.myhuaweicloud.com/dmetasoul-repo/rustfs/rustfs:1.0.0-beta.3 | |
| ports: | |
| - "9000:9000" | |
| - "9001:9001" | |
| env: | |
| RUSTFS_ACCESS_KEY: rustfsadmin | |
| RUSTFS_SECRET_KEY: rustfsadmin | |
| RUSTFS_CONSOLE_ENABLE: true | |
| steps: | |
| - name: Wait for RustFS | |
| run: | | |
| for i in {1..60}; do | |
| if curl -fsS http://127.0.0.1:9000/health \ | |
| && curl -fsS http://127.0.0.1:9001/rustfs/console/health; then | |
| echo "RustFS is ready" | |
| exit 0 | |
| fi | |
| echo "Waiting for RustFS..." | |
| sleep 2 | |
| done | |
| echo "RustFS failed to become ready" | |
| exit 1 | |
| - name: Setup RustFS bucket | |
| run: | | |
| docker run --rm --network host \ | |
| --entrypoint /bin/sh \ | |
| swr.cn-southwest-2.myhuaweicloud.com/dmetasoul-repo/rustfs/rc:v0.1.16 \ | |
| -c ' | |
| set -e | |
| until rc alias set rustfs http://127.0.0.1:9000 rustfsadmin rustfsadmin; do | |
| echo "RustFS is not ready yet, retrying in 2 seconds..." | |
| sleep 2 | |
| done | |
| until rc mb rustfs/lakesoul-test-bucket --ignore-existing; do | |
| echo "Bucket creation failed, retrying in 2 seconds..." | |
| sleep 2 | |
| done | |
| ' | |
| - name: Free Disk Space (Ubuntu) | |
| uses: jlumbroso/free-disk-space@main | |
| with: | |
| tool-cache: false | |
| android: true | |
| dotnet: true | |
| haskell: true | |
| large-packages: false | |
| docker-images: true | |
| swap-storage: true | |
| - uses: actions/checkout@v4 | |
| - name: Set up JDK 11 | |
| uses: actions/setup-java@v4 | |
| with: | |
| java-version: "11" | |
| distribution: "temurin" | |
| cache: maven | |
| - name: Set up Python 3.10 | |
| uses: actions/setup-python@v5 | |
| with: | |
| python-version: "3.10" | |
| - name: Install Rust toolchain | |
| uses: dtolnay/rust-toolchain@master | |
| with: | |
| toolchain: stable | |
| - name: Install Protoc | |
| uses: arduino/setup-protoc@v2 | |
| with: | |
| version: "23.x" | |
| repo-token: ${{ secrets.GITHUB_TOKEN }} | |
| - name: Install uv | |
| uses: astral-sh/setup-uv@v6 | |
| with: | |
| enable-cache: true | |
| cache-suffix: "lakesoul-compat-ci" | |
| cache-dependency-glob: | | |
| python/pyproject.toml | |
| python/uv.lock | |
| - uses: Swatinem/rust-cache@v2 | |
| with: | |
| workspaces: ". -> rust/target" | |
| env-vars: "JAVA_HOME" | |
| - name: Install psql | |
| run: sudo apt-get install -y postgresql-client-16 | |
| - name: Init PostgreSQL metadata | |
| run: ./script/meta_init_for_local_test.sh -j 1 | |
| - name: Build native libraries | |
| run: cargo -q build --release -p lakesoul-io-c -p lakesoul-metadata-c | |
| - name: Build Spark and Flink test artifacts | |
| run: mvn -q -B package -pl lakesoul-spark,lakesoul-flink -am -Pcross-build -DskipTests --file pom.xml | |
| - name: Build Python package | |
| working-directory: python | |
| run: | | |
| uv sync --python 3.10 --extra pandas --extra ray --extra daft --extra pyspark | |
| - name: Run RustFS compatibility matrix | |
| env: | |
| LAKESOUL_SOURCE_DIR: ${{ github.workspace }} | |
| LAKESOUL_COMPAT_S3_BUCKET: lakesoul-test-bucket | |
| LAKESOUL_COMPAT_S3_ENDPOINT: http://127.0.0.1:9000 | |
| LAKESOUL_COMPAT_S3_ACCESS_KEY: rustfsadmin | |
| LAKESOUL_COMPAT_S3_SECRET_KEY: rustfsadmin | |
| PYTHONPATH: tests:src | |
| working-directory: python | |
| run: | | |
| EVENT_NAME="${{ github.event_name }}" | |
| if [ "$EVENT_NAME" = "workflow_dispatch" ]; then | |
| MODE="${{ inputs.cases }}" | |
| ENGINES="${{ inputs.engines }}" | |
| else | |
| MODE="full" | |
| ENGINES="all" | |
| fi | |
| .venv/bin/python -m compat.run_matrix \ | |
| --mode "$MODE" \ | |
| --cases "$MODE" \ | |
| --engines "$ENGINES" \ | |
| --storage "s3://lakesoul-test-bucket/compat" \ | |
| --output-dir "${{ github.workspace }}/compatibility-artifacts/rustfs" | |
| - name: Dump PostgreSQL metadata | |
| if: always() | |
| run: | | |
| mkdir -p compatibility-artifacts/rustfs/postgres | |
| PGPASSWORD=lakesoul_test psql -h 127.0.0.1 -p 5432 -U lakesoul_test -d lakesoul_test -c "select * from table_info;" > compatibility-artifacts/rustfs/postgres/table_info.txt | |
| PGPASSWORD=lakesoul_test psql -h 127.0.0.1 -p 5432 -U lakesoul_test -d lakesoul_test -c "select * from data_commit_info;" > compatibility-artifacts/rustfs/postgres/data_commit_info.txt | |
| - name: Upload compatibility artifacts | |
| if: always() | |
| uses: actions/upload-artifact@v4 | |
| with: | |
| name: compatibility-rustfs-artifacts | |
| path: compatibility-artifacts/rustfs | |
| retention-days: 7 | |
| if-no-files-found: warn |