ci: move components to flakes so it won't affect builders (#5464)

* ci: move components to flakes so it won't affect builders

* chore: add gnuplot for benchmark/criterion

.github/actions/build-images/action.yml

@@ -41,8 +41,8 @@ runs:
         image-name: ${{ inputs.image-name }}
         image-tag: ${{ inputs.version }}
         docker-file: docker/ci/ubuntu/Dockerfile
-        amd64-artifact-name: greptime-linux-amd64-pyo3-${{ inputs.version }}
-        arm64-artifact-name: greptime-linux-arm64-pyo3-${{ inputs.version }}
+        amd64-artifact-name: greptime-linux-amd64-${{ inputs.version }}
+        arm64-artifact-name: greptime-linux-arm64-${{ inputs.version }}
         platforms: linux/amd64,linux/arm64
         push-latest-tag: ${{ inputs.push-latest-tag }}
 

.github/actions/build-linux-artifacts/action.yml

@@ -48,24 +48,11 @@ runs:
         path: /tmp/greptime-*.log
         retention-days: 3
 
-    - name: Build standard greptime
+    - name: Build greptime # Builds standard greptime binary
       uses: ./.github/actions/build-greptime-binary
       with:
         base-image: ubuntu
-        features: pyo3_backend,servers/dashboard
-        cargo-profile: ${{ inputs.cargo-profile }}
-        artifacts-dir: greptime-linux-${{ inputs.arch }}-pyo3-${{ inputs.version }}
-        version: ${{ inputs.version }}
-        working-dir: ${{ inputs.working-dir }}
-        image-registry: ${{ inputs.image-registry }}
-        image-namespace: ${{ inputs.image-namespace }}
-
-    - name: Build greptime without pyo3
-      if: ${{ inputs.dev-mode == 'false' }}
-      uses: ./.github/actions/build-greptime-binary
-      with:
-        base-image: ubuntu
-        features: servers/dashboard
+        features: servers/dashboard,pg_kvbackend
         cargo-profile: ${{ inputs.cargo-profile }}
         artifacts-dir: greptime-linux-${{ inputs.arch }}-${{ inputs.version }}
         version: ${{ inputs.version }}
@@ -83,7 +70,7 @@ runs:
       if: ${{ inputs.arch == 'amd64' && inputs.dev-mode == 'false' }} # Builds greptime for centos if the host machine is amd64.
       with:
         base-image: centos
-        features: servers/dashboard
+        features: servers/dashboard,pg_kvbackend
         cargo-profile: ${{ inputs.cargo-profile }}
         artifacts-dir: greptime-linux-${{ inputs.arch }}-centos-${{ inputs.version }}
         version: ${{ inputs.version }}

.github/actions/build-windows-artifacts/action.yml

@@ -33,15 +33,6 @@ runs:
     - name: Rust Cache
       uses: Swatinem/rust-cache@v2
 
-    - name: Install Python
-      uses: actions/setup-python@v5
-      with:
-        python-version: "3.10"
-
-    - name: Install PyArrow Package
-      shell: pwsh
-      run: pip install pyarrow numpy
-
     - name: Install WSL distribution
       uses: Vampire/setup-wsl@v2
       with:
@@ -76,5 +67,5 @@ runs:
       uses: ./.github/actions/upload-artifacts
       with:
         artifacts-dir: ${{ inputs.artifacts-dir }}
-        target-files: target/${{ inputs.arch }}/${{ inputs.cargo-profile }}/greptime
+        target-files: target/${{ inputs.arch }}/${{ inputs.cargo-profile }}/greptime,target/${{ inputs.arch }}/${{ inputs.cargo-profile }}/greptime.pdb
         version: ${{ inputs.version }}

.github/actions/publish-github-release/action.yml

@@ -9,8 +9,8 @@ runs:
   steps:
     # Download artifacts from previous jobs, the artifacts will be downloaded to:
     # ${WORKING_DIR}
-    #   |- greptime-darwin-amd64-pyo3-v0.5.0/greptime-darwin-amd64-pyo3-v0.5.0.tar.gz
-    #   |- greptime-darwin-amd64-pyo3-v0.5.0.sha256sum/greptime-darwin-amd64-pyo3-v0.5.0.sha256sum
+    #   |- greptime-darwin-amd64-v0.5.0/greptime-darwin-amd64-v0.5.0.tar.gz
+    #   |- greptime-darwin-amd64-v0.5.0.sha256sum/greptime-darwin-amd64-v0.5.0.sha256sum
     #   |- greptime-darwin-amd64-v0.5.0/greptime-darwin-amd64-v0.5.0.tar.gz
     #   |- greptime-darwin-amd64-v0.5.0.sha256sum/greptime-darwin-amd64-v0.5.0.sha256sum
     #   ...

.github/actions/upload-artifacts/action.yml

@@ -30,9 +30,9 @@ runs:
         done
 
     # The compressed artifacts will use the following layout:
-    # greptime-linux-amd64-pyo3-v0.3.0sha256sum
-    # greptime-linux-amd64-pyo3-v0.3.0.tar.gz
-    #   greptime-linux-amd64-pyo3-v0.3.0
+    # greptime-linux-amd64-v0.3.0sha256sum
+    # greptime-linux-amd64-v0.3.0.tar.gz
+    #   greptime-linux-amd64-v0.3.0
     #   └── greptime
     - name: Compress artifacts and calculate checksum
       working-directory: ${{ inputs.working-dir }}

.github/scripts/upload-artifacts-to-s3.sh

@@ -27,11 +27,11 @@ function upload_artifacts() {
   # ├── latest-version.txt
   # ├── latest-nightly-version.txt
   # ├── v0.1.0
-  # │   ├── greptime-darwin-amd64-pyo3-v0.1.0.sha256sum
-  # │   └── greptime-darwin-amd64-pyo3-v0.1.0.tar.gz
+  # │   ├── greptime-darwin-amd64-v0.1.0.sha256sum
+  # │   └── greptime-darwin-amd64-v0.1.0.tar.gz
   # └── v0.2.0
-  #    ├── greptime-darwin-amd64-pyo3-v0.2.0.sha256sum
-  #    └── greptime-darwin-amd64-pyo3-v0.2.0.tar.gz
+  #    ├── greptime-darwin-amd64-v0.2.0.sha256sum
+  #    └── greptime-darwin-amd64-v0.2.0.tar.gz
   find "$ARTIFACTS_DIR" -type f \( -name "*.tar.gz" -o -name "*.sha256sum" \) | while IFS= read -r file; do
     aws s3 cp \
       "$file" "s3://$AWS_S3_BUCKET/$RELEASE_DIRS/$VERSION/$(basename "$file")"

.github/workflows/dependency-check.yml

@@ -1,9 +1,6 @@
 name: Check Dependencies
 
 on:
-  push:
-    branches:
-      - main
   pull_request:
     branches:
       - main

.github/workflows/develop.yml

@@ -1,4 +1,6 @@
 on:
+  schedule:
+    - cron: "0 15 * * 1-5"
   merge_group:
   pull_request:
     types: [ opened, synchronize, reopened, ready_for_review ]
@@ -10,17 +12,6 @@ on:
       - 'docker/**'
       - '.gitignore'
       - 'grafana/**'
-  push:
-    branches:
-      - main
-    paths-ignore:
-      - 'docs/**'
-      - 'config/**'
-      - '**.md'
-      - '.dockerignore'
-      - 'docker/**'
-      - '.gitignore'
-      - 'grafana/**'
   workflow_dispatch:
 
 name: CI
@@ -54,7 +45,7 @@ jobs:
     runs-on: ${{ matrix.os }}
     strategy:
       matrix:
-        os: [ windows-2022, ubuntu-20.04 ]
+        os: [ ubuntu-20.04 ]
     timeout-minutes: 60
     steps:
       - uses: actions/checkout@v4
@@ -68,6 +59,8 @@ jobs:
           # Shares across multiple jobs
           # Shares with `Clippy` job
           shared-key: "check-lint"
+          cache-all-crates: "true"
+          save-if: ${{ github.ref == 'refs/heads/main' }}
       - name: Run cargo check
         run: cargo check --locked --workspace --all-targets
 
@@ -78,13 +71,8 @@ jobs:
     steps:
       - uses: actions/checkout@v4
       - uses: actions-rust-lang/setup-rust-toolchain@v1
-      - name: Rust Cache
-        uses: Swatinem/rust-cache@v2
-        with:
-          # Shares across multiple jobs
-          shared-key: "check-toml"
       - name: Install taplo
-        run: cargo +stable install taplo-cli --version ^0.9 --locked
+        run: cargo +stable install taplo-cli --version ^0.9 --locked --force
       - name: Run taplo
         run: taplo format --check
 
@@ -105,13 +93,15 @@ jobs:
         with:
           # Shares across multiple jobs
           shared-key: "build-binaries"
+          cache-all-crates: "true"
+          save-if: ${{ github.ref == 'refs/heads/main' }}
       - name: Install cargo-gc-bin
         shell: bash
-        run: cargo install cargo-gc-bin
+        run: cargo install cargo-gc-bin --force
       - name: Build greptime binaries
         shell: bash
         # `cargo gc` will invoke `cargo build` with specified args
-        run: cargo gc -- --bin greptime --bin sqlness-runner
+        run: cargo gc -- --bin greptime --bin sqlness-runner --features pg_kvbackend
       - name: Pack greptime binaries
         shell: bash
         run: |
@@ -153,17 +143,12 @@ jobs:
         with:
           repo-token: ${{ secrets.GITHUB_TOKEN }}
       - uses: actions-rust-lang/setup-rust-toolchain@v1
-      - name: Rust Cache
-        uses: Swatinem/rust-cache@v2
-        with:
-          # Shares across multiple jobs
-          shared-key: "fuzz-test-targets"
       - name: Set Rust Fuzz
         shell: bash
         run: |
           sudo apt-get install -y libfuzzer-14-dev
           rustup install nightly
-          cargo +nightly install cargo-fuzz cargo-gc-bin
+          cargo +nightly install cargo-fuzz cargo-gc-bin --force
       - name: Download pre-built binaries
         uses: actions/download-artifact@v4
         with:
@@ -211,16 +196,11 @@ jobs:
         with:
           repo-token: ${{ secrets.GITHUB_TOKEN }}
       - uses: actions-rust-lang/setup-rust-toolchain@v1
-      - name: Rust Cache
-        uses: Swatinem/rust-cache@v2
-        with:
-          # Shares across multiple jobs
-          shared-key: "fuzz-test-targets"
       - name: Set Rust Fuzz
         shell: bash
         run: |
           sudo apt update && sudo apt install -y libfuzzer-14-dev
-          cargo install cargo-fuzz cargo-gc-bin
+          cargo install cargo-fuzz cargo-gc-bin --force
       - name: Download pre-built binariy
         uses: actions/download-artifact@v4
         with:
@@ -266,13 +246,15 @@ jobs:
         with:
           # Shares across multiple jobs
           shared-key: "build-greptime-ci"
+          cache-all-crates: "true"
+          save-if: ${{ github.ref == 'refs/heads/main' }}
       - name: Install cargo-gc-bin
         shell: bash
-        run: cargo install cargo-gc-bin
+        run: cargo install cargo-gc-bin --force
       - name: Build greptime bianry
         shell: bash
         # `cargo gc` will invoke `cargo build` with specified args
-        run: cargo gc --profile ci -- --bin greptime
+        run: cargo gc --profile ci -- --bin greptime --features pg_kvbackend
       - name: Pack greptime binary
         shell: bash
         run: |
@@ -328,17 +310,12 @@ jobs:
         with:
           repo-token: ${{ secrets.GITHUB_TOKEN }}
       - uses: actions-rust-lang/setup-rust-toolchain@v1
-      - name: Rust Cache
-        uses: Swatinem/rust-cache@v2
-        with:
-          # Shares across multiple jobs
-          shared-key: "fuzz-test-targets"
       - name: Set Rust Fuzz
         shell: bash
         run: |
           sudo apt-get install -y libfuzzer-14-dev
           rustup install nightly
-          cargo +nightly install cargo-fuzz cargo-gc-bin
+          cargo +nightly install cargo-fuzz cargo-gc-bin --force
       # Downloads ci image
       - name: Download pre-built binariy
         uses: actions/download-artifact@v4
@@ -477,17 +454,12 @@ jobs:
         with:
           repo-token: ${{ secrets.GITHUB_TOKEN }}
       - uses: actions-rust-lang/setup-rust-toolchain@v1
-      - name: Rust Cache
-        uses: Swatinem/rust-cache@v2
-        with:
-          # Shares across multiple jobs
-          shared-key: "fuzz-test-targets"
       - name: Set Rust Fuzz
         shell: bash
         run: |
           sudo apt-get install -y libfuzzer-14-dev
           rustup install nightly
-          cargo +nightly install cargo-fuzz cargo-gc-bin
+          cargo +nightly install cargo-fuzz cargo-gc-bin --force
       # Downloads ci image
       - name: Download pre-built binariy
         uses: actions/download-artifact@v4
@@ -584,13 +556,16 @@ jobs:
           - name: "Remote WAL"
             opts: "-w kafka -k 127.0.0.1:9092"
             kafka: true
+          - name: "Pg Kvbackend"
+            opts: "--setup-pg"
+            kafka: false
     timeout-minutes: 60
     steps:
       - uses: actions/checkout@v4
       - if: matrix.mode.kafka
         name: Setup kafka server
-        working-directory: tests-integration/fixtures/kafka
-        run: docker compose -f docker-compose-standalone.yml up -d --wait
+        working-directory: tests-integration/fixtures
+        run: docker compose up -d --wait kafka
       - name: Download pre-built binaries
         uses: actions/download-artifact@v4
         with:
@@ -620,11 +595,6 @@ jobs:
       - uses: actions-rust-lang/setup-rust-toolchain@v1
         with:
           components: rustfmt
-      - name: Rust Cache
-        uses: Swatinem/rust-cache@v2
-        with:
-          # Shares across multiple jobs
-          shared-key: "check-rust-fmt"
       - name: Check format
         run: make fmt-check
 
@@ -646,11 +616,70 @@ jobs:
           # Shares across multiple jobs
           # Shares with `Check` job
           shared-key: "check-lint"
+          cache-all-crates: "true"
+          save-if: ${{ github.ref == 'refs/heads/main' }}
       - name: Run cargo clippy
         run: make clippy
 
+  conflict-check:
+    name: Check for conflict
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+      - name: Merge Conflict Finder
+        uses: olivernybroe/action-conflict-finder@v4.0
+
+  test:
+    if: github.event_name != 'merge_group'
+    runs-on: ubuntu-22.04-arm
+    timeout-minutes: 60
+    needs:  [conflict-check, clippy, fmt]
+    steps:
+      - uses: actions/checkout@v4
+      - uses: arduino/setup-protoc@v3
+        with:
+          repo-token: ${{ secrets.GITHUB_TOKEN }}
+      - uses: rui314/setup-mold@v1
+      - name: Install toolchain
+        uses: actions-rust-lang/setup-rust-toolchain@v1
+        with:
+            cache: false
+      - name: Rust Cache
+        uses: Swatinem/rust-cache@v2
+        with:
+          # Shares cross multiple jobs
+          shared-key: "coverage-test"
+          cache-all-crates: "true"
+          save-if: ${{ github.ref == 'refs/heads/main' }}
+      - name: Install latest nextest release
+        uses: taiki-e/install-action@nextest
+      - name: Setup external services
+        working-directory: tests-integration/fixtures
+        run: docker compose up -d --wait
+      - name: Run nextest cases
+        run: cargo nextest run --workspace -F dashboard -F pg_kvbackend
+        env:
+          CARGO_BUILD_RUSTFLAGS: "-C link-arg=-fuse-ld=mold"
+          RUST_BACKTRACE: 1
+          RUST_MIN_STACK: 8388608 # 8MB
+          CARGO_INCREMENTAL: 0
+          GT_S3_BUCKET: ${{ vars.AWS_CI_TEST_BUCKET }}
+          GT_S3_ACCESS_KEY_ID: ${{ secrets.AWS_CI_TEST_ACCESS_KEY_ID }}
+          GT_S3_ACCESS_KEY: ${{ secrets.AWS_CI_TEST_SECRET_ACCESS_KEY }}
+          GT_S3_REGION: ${{ vars.AWS_CI_TEST_BUCKET_REGION }}
+          GT_MINIO_BUCKET: greptime
+          GT_MINIO_ACCESS_KEY_ID: superpower_ci_user
+          GT_MINIO_ACCESS_KEY: superpower_password
+          GT_MINIO_REGION: us-west-2
+          GT_MINIO_ENDPOINT_URL: http://127.0.0.1:9000
+          GT_ETCD_ENDPOINTS: http://127.0.0.1:2379
+          GT_POSTGRES_ENDPOINTS: postgres://greptimedb:admin@127.0.0.1:5432/postgres
+          GT_KAFKA_ENDPOINTS: 127.0.0.1:9092
+          GT_KAFKA_SASL_ENDPOINTS: 127.0.0.1:9093
+          UNITTEST_LOG_DIR: "__unittest_logs"
+
   coverage:
-    if: github.event.pull_request.draft == false
+    if: github.event_name == 'merge_group'
     runs-on: ubuntu-20.04-8-cores
     timeout-minutes: 60
     steps:
@@ -658,48 +687,29 @@ jobs:
       - uses: arduino/setup-protoc@v3
         with:
           repo-token: ${{ secrets.GITHUB_TOKEN }}
-      - uses: KyleMayes/install-llvm-action@v1
-        with:
-          version: "14.0"
+      - uses: rui314/setup-mold@v1
       - name: Install toolchain
         uses: actions-rust-lang/setup-rust-toolchain@v1
         with:
-          components: llvm-tools-preview
+          components: llvm-tools
+          cache: false
       - name: Rust Cache
         uses: Swatinem/rust-cache@v2
         with:
           # Shares cross multiple jobs
           shared-key: "coverage-test"
-      - name: Docker Cache
-        uses: ScribeMD/docker-cache@0.3.7
-        with:
-          key: docker-${{ runner.os }}-coverage
+          save-if: ${{ github.ref == 'refs/heads/main' }}
       - name: Install latest nextest release
         uses: taiki-e/install-action@nextest
       - name: Install cargo-llvm-cov
         uses: taiki-e/install-action@cargo-llvm-cov
-      - name: Install Python
-        uses: actions/setup-python@v5
-        with:
-          python-version: '3.10'
-      - name: Install PyArrow Package
-        run: pip install pyarrow numpy
-      - name: Setup etcd server
-        working-directory: tests-integration/fixtures/etcd
-        run: docker compose -f docker-compose-standalone.yml up -d --wait
-      - name: Setup kafka server
-        working-directory: tests-integration/fixtures/kafka
-        run: docker compose -f docker-compose-standalone.yml up -d --wait
-      - name: Setup minio
-        working-directory: tests-integration/fixtures/minio
-        run: docker compose -f docker-compose-standalone.yml up -d --wait
-      - name: Setup postgres server
-        working-directory: tests-integration/fixtures/postgres
-        run: docker compose -f docker-compose-standalone.yml up -d --wait
+      - name: Setup external services
+        working-directory: tests-integration/fixtures
+        run: docker compose up -d --wait
       - name: Run nextest cases
-        run: cargo llvm-cov nextest --workspace --lcov --output-path lcov.info -F pyo3_backend -F dashboard
+        run: cargo llvm-cov nextest --workspace --lcov --output-path lcov.info -F dashboard -F pg_kvbackend
         env:
-          CARGO_BUILD_RUSTFLAGS: "-C link-arg=-fuse-ld=lld"
+          CARGO_BUILD_RUSTFLAGS: "-C link-arg=-fuse-ld=mold"
           RUST_BACKTRACE: 1
           CARGO_INCREMENTAL: 0
           GT_S3_BUCKET: ${{ vars.AWS_CI_TEST_BUCKET }}

.github/workflows/docs.yml

@@ -66,6 +66,11 @@ jobs:
     steps:
       - run: 'echo "No action required"'
 
+  test:
+    runs-on: ubuntu-20.04
+    steps:
+      - run: 'echo "No action required"'
+
   sqlness:
     name: Sqlness Test (${{ matrix.mode.name }})
     runs-on: ${{ matrix.os }}

.github/workflows/nightly-ci.yml

@@ -1,6 +1,6 @@
 on:
   schedule:
-    - cron: "0 23 * * 1-5"
+    - cron: "0 23 * * 1-4"
   workflow_dispatch:
 
 name: Nightly CI
@@ -91,18 +91,12 @@ jobs:
         uses: Swatinem/rust-cache@v2
       - name: Install Cargo Nextest
         uses: taiki-e/install-action@nextest
-      - name: Install Python
-        uses: actions/setup-python@v5
-        with:
-          python-version: "3.10"
-      - name: Install PyArrow Package
-        run: pip install pyarrow numpy
       - name: Install WSL distribution
         uses: Vampire/setup-wsl@v2
         with:
           distribution: Ubuntu-22.04
       - name: Running tests
-        run: cargo nextest run -F pyo3_backend,dashboard
+        run: cargo nextest run -F dashboard
         env:
           CARGO_BUILD_RUSTFLAGS: "-C linker=lld-link"
           RUST_BACKTRACE: 1
@@ -115,14 +109,15 @@ jobs:
           UNITTEST_LOG_DIR: "__unittest_logs"
 
   cleanbuild-linux-nix:
-    runs-on: ubuntu-latest-8-cores
+    name: Run clean build on Linux
+    runs-on: ubuntu-latest
     timeout-minutes: 60
     steps:
       - uses: actions/checkout@v4
       - uses: cachix/install-nix-action@v27
         with:
-          nix_path: nixpkgs=channel:nixos-unstable
-      - run: nix-shell --pure --run "cargo build"
+          nix_path: nixpkgs=channel:nixos-24.11
+      - run: nix develop --command cargo build
 
   check-status:
     name: Check status

.github/workflows/release.yml

@@ -222,18 +222,10 @@ jobs:
             arch: aarch64-apple-darwin
             features: servers/dashboard
             artifacts-dir-prefix: greptime-darwin-arm64
-          - os: ${{ needs.allocate-runners.outputs.macos-runner }}
-            arch: aarch64-apple-darwin
-            features: pyo3_backend,servers/dashboard
-            artifacts-dir-prefix: greptime-darwin-arm64-pyo3
           - os: ${{ needs.allocate-runners.outputs.macos-runner }}
             features: servers/dashboard
             arch: x86_64-apple-darwin
             artifacts-dir-prefix: greptime-darwin-amd64
-          - os: ${{ needs.allocate-runners.outputs.macos-runner }}
-            features: pyo3_backend,servers/dashboard
-            arch: x86_64-apple-darwin
-            artifacts-dir-prefix: greptime-darwin-amd64-pyo3
     runs-on: ${{ matrix.os }}
     outputs:
       build-macos-result: ${{ steps.set-build-macos-result.outputs.build-macos-result }}
@@ -271,10 +263,6 @@ jobs:
             arch: x86_64-pc-windows-msvc
             features: servers/dashboard
             artifacts-dir-prefix: greptime-windows-amd64
-          - os: ${{ needs.allocate-runners.outputs.windows-runner }}
-            arch: x86_64-pc-windows-msvc
-            features: pyo3_backend,servers/dashboard
-            artifacts-dir-prefix: greptime-windows-amd64-pyo3
     runs-on: ${{ matrix.os }}
     outputs:
       build-windows-result: ${{ steps.set-build-windows-result.outputs.build-windows-result }}
@@ -448,6 +436,22 @@ jobs:
           aws-region: ${{ vars.EC2_RUNNER_REGION }}
           github-token: ${{ secrets.GH_PERSONAL_ACCESS_TOKEN }}
 
+  bump-doc-version:
+    name: Bump doc version
+    if: ${{ github.event_name == 'push' || github.event_name == 'schedule' }}
+    needs: [allocate-runners]
+    runs-on: ubuntu-20.04
+    steps:
+      - uses: actions/checkout@v4
+      - uses: ./.github/actions/setup-cyborg
+      - name: Bump doc version
+        working-directory: cyborg
+        run: pnpm tsx bin/bump-doc-version.ts
+        env:
+          VERSION: ${{ needs.allocate-runners.outputs.version }}
+          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+          DOCS_REPO_TOKEN: ${{ secrets.DOCS_REPO_TOKEN }}
+
   notification:
     if: ${{ github.repository == 'GreptimeTeam/greptimedb' && (github.event_name == 'push' || github.event_name == 'schedule') && always() }}
     name: Send notification to Greptime team

Cargo.toml

@@ -55,7 +55,6 @@ members = [
     "src/promql",
     "src/puffin",
     "src/query",
-    "src/script",
     "src/servers",
     "src/session",
     "src/sql",
@@ -79,8 +78,6 @@ clippy.dbg_macro = "warn"
 clippy.implicit_clone = "warn"
 clippy.readonly_write_lock = "allow"
 rust.unknown_lints = "deny"
-# Remove this after https://github.com/PyO3/pyo3/issues/4094
-rust.non_local_definitions = "allow"
 rust.unexpected_cfgs = { level = "warn", check-cfg = ['cfg(tokio_unstable)'] }
 
 [workspace.dependencies]
@@ -91,14 +88,18 @@ rust.unexpected_cfgs = { level = "warn", check-cfg = ['cfg(tokio_unstable)'] }
 # See for more detaiils: https://github.com/rust-lang/cargo/issues/11329
 ahash = { version = "0.8", features = ["compile-time-rng"] }
 aquamarine = "0.3"
-arrow = { version = "51.0.0", features = ["prettyprint"] }
-arrow-array = { version = "51.0.0", default-features = false, features = ["chrono-tz"] }
-arrow-flight = "51.0"
-arrow-ipc = { version = "51.0.0", default-features = false, features = ["lz4", "zstd"] }
-arrow-schema = { version = "51.0", features = ["serde"] }
+arrow = { version = "53.0.0", features = ["prettyprint"] }
+arrow-array = { version = "53.0.0", default-features = false, features = ["chrono-tz"] }
+arrow-flight = "53.0"
+arrow-ipc = { version = "53.0.0", default-features = false, features = ["lz4", "zstd"] }
+arrow-schema = { version = "53.0", features = ["serde"] }
 async-stream = "0.3"
 async-trait = "0.1"
-axum = { version = "0.6", features = ["headers"] }
+# Remember to update axum-extra, axum-macros when updating axum
+axum = "0.8"
+axum-extra = "0.10"
+axum-macros = "0.4"
+backon = "1"
 base64 = "0.21"
 bigdecimal = "0.4.2"
 bitflags = "2.4.1"
@@ -109,35 +110,43 @@ clap = { version = "4.4", features = ["derive"] }
 config = "0.13.0"
 crossbeam-utils = "0.8"
 dashmap = "5.4"
-datafusion = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "7823ef2f63663907edab46af0d51359900f608d6" }
-datafusion-common = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "7823ef2f63663907edab46af0d51359900f608d6" }
-datafusion-expr = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "7823ef2f63663907edab46af0d51359900f608d6" }
-datafusion-functions = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "7823ef2f63663907edab46af0d51359900f608d6" }
-datafusion-optimizer = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "7823ef2f63663907edab46af0d51359900f608d6" }
-datafusion-physical-expr = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "7823ef2f63663907edab46af0d51359900f608d6" }
-datafusion-physical-plan = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "7823ef2f63663907edab46af0d51359900f608d6" }
-datafusion-sql = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "7823ef2f63663907edab46af0d51359900f608d6" }
-datafusion-substrait = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "7823ef2f63663907edab46af0d51359900f608d6" }
+datafusion = { git = "https://github.com/apache/datafusion.git", rev = "2464703c84c400a09cc59277018813f0e797bb4e" }
+datafusion-common = { git = "https://github.com/apache/datafusion.git", rev = "2464703c84c400a09cc59277018813f0e797bb4e" }
+datafusion-expr = { git = "https://github.com/apache/datafusion.git", rev = "2464703c84c400a09cc59277018813f0e797bb4e" }
+datafusion-functions = { git = "https://github.com/apache/datafusion.git", rev = "2464703c84c400a09cc59277018813f0e797bb4e" }
+datafusion-optimizer = { git = "https://github.com/apache/datafusion.git", rev = "2464703c84c400a09cc59277018813f0e797bb4e" }
+datafusion-physical-expr = { git = "https://github.com/apache/datafusion.git", rev = "2464703c84c400a09cc59277018813f0e797bb4e" }
+datafusion-physical-plan = { git = "https://github.com/apache/datafusion.git", rev = "2464703c84c400a09cc59277018813f0e797bb4e" }
+datafusion-sql = { git = "https://github.com/apache/datafusion.git", rev = "2464703c84c400a09cc59277018813f0e797bb4e" }
+datafusion-substrait = { git = "https://github.com/apache/datafusion.git", rev = "2464703c84c400a09cc59277018813f0e797bb4e" }
+deadpool = "0.10"
+deadpool-postgres = "0.12"
 derive_builder = "0.12"
 dotenv = "0.15"
-etcd-client = "0.13"
+etcd-client = "0.14"
 fst = "0.4.7"
 futures = "0.3"
 futures-util = "0.3"
-greptime-proto = { git = "https://github.com/GreptimeTeam/greptime-proto.git", rev = "a875e976441188028353f7274a46a7e6e065c5d4" }
+greptime-proto = { git = "https://github.com/GreptimeTeam/greptime-proto.git", rev = "683e9d10ae7f3dfb8aaabd89082fc600c17e3795" }
 hex = "0.4"
+http = "1"
 humantime = "2.1"
 humantime-serde = "1.1"
+hyper = "1.1"
+hyper-util = "0.1"
 itertools = "0.10"
 jsonb = { git = "https://github.com/databendlabs/jsonb.git", rev = "8c8d2fc294a39f3ff08909d60f718639cfba3875", default-features = false }
 lazy_static = "1.4"
+local-ip-address = "0.6"
+loki-api = { git = "https://github.com/shuiyisong/tracing-loki", branch = "chore/prost_version" }
 meter-core = { git = "https://github.com/GreptimeTeam/greptime-meter.git", rev = "a10facb353b41460eeb98578868ebf19c2084fac" }
 mockall = "0.11.4"
 moka = "0.12"
+nalgebra = "0.33"
 notify = "6.1"
 num_cpus = "1.16"
 once_cell = "1.18"
-opentelemetry-proto = { version = "0.5", features = [
+opentelemetry-proto = { version = "0.27", features = [
     "gen-tonic",
     "metrics",
     "trace",
@@ -145,12 +154,12 @@ opentelemetry-proto = { version = "0.5", features = [
     "logs",
 ] }
 parking_lot = "0.12"
-parquet = { version = "51.0.0", default-features = false, features = ["arrow", "async", "object_store"] }
+parquet = { version = "53.0.0", default-features = false, features = ["arrow", "async", "object_store"] }
 paste = "1.0"
 pin-project = "1.0"
 prometheus = { version = "0.13.3", features = ["process"] }
 promql-parser = { version = "0.4.3", features = ["ser"] }
-prost = "0.12"
+prost = "0.13"
 raft-engine = { version = "0.4.1", default-features = false }
 rand = "0.8"
 ratelimit = "0.9"
@@ -169,28 +178,30 @@ rstest = "0.21"
 rstest_reuse = "0.7"
 rust_decimal = "1.33"
 rustc-hash = "2.0"
+rustls = { version = "0.23.20", default-features = false } # override by patch, see [patch.crates-io]
 serde = { version = "1.0", features = ["derive"] }
 serde_json = { version = "1.0", features = ["float_roundtrip"] }
 serde_with = "3"
-shadow-rs = "0.35"
+shadow-rs = "0.38"
 similar-asserts = "1.6.0"
 smallvec = { version = "1", features = ["serde"] }
 snafu = "0.8"
 sysinfo = "0.30"
-# on branch v0.44.x
-sqlparser = { git = "https://github.com/GreptimeTeam/sqlparser-rs.git", rev = "54a267ac89c09b11c0c88934690530807185d3e7", features = [
+# on branch v0.52.x
+sqlparser = { git = "https://github.com/GreptimeTeam/sqlparser-rs.git", rev = "71dd86058d2af97b9925093d40c4e03360403170", features = [
     "visitor",
     "serde",
-] }
+] } # on branch v0.44.x
 strum = { version = "0.25", features = ["derive"] }
 tempfile = "3"
 tokio = { version = "1.40", features = ["full"] }
 tokio-postgres = "0.7"
+tokio-rustls = { version = "0.26.0", default-features = false } # override by patch, see [patch.crates-io]
 tokio-stream = "0.1"
 tokio-util = { version = "0.7", features = ["io-util", "compat"] }
 toml = "0.8.8"
-tonic = { version = "0.11", features = ["tls", "gzip", "zstd"] }
-tower = "0.4"
+tonic = { version = "0.12", features = ["tls", "gzip", "zstd"] }
+tower = "0.5"
 tracing-appender = "0.2"
 tracing-subscriber = { version = "0.3", features = ["env-filter", "json", "fmt"] }
 typetag = "0.2"
@@ -252,7 +263,6 @@ plugins = { path = "src/plugins" }
 promql = { path = "src/promql" }
 puffin = { path = "src/puffin" }
 query = { path = "src/query" }
-script = { path = "src/script" }
 servers = { path = "src/servers" }
 session = { path = "src/session" }
 sql = { path = "src/sql" }
@@ -262,9 +272,9 @@ table = { path = "src/table" }
 
 [patch.crates-io]
 # change all rustls dependencies to use our fork to default to `ring` to make it "just work"
-hyper-rustls = { git = "https://github.com/GreptimeTeam/hyper-rustls" }
-rustls = { git = "https://github.com/GreptimeTeam/rustls" }
-tokio-rustls = { git = "https://github.com/GreptimeTeam/tokio-rustls" }
+hyper-rustls = { git = "https://github.com/GreptimeTeam/hyper-rustls", rev = "a951e03" } # version = "0.27.5" with ring patch
+rustls = { git = "https://github.com/GreptimeTeam/rustls", rev = "34fd0c6" }             # version = "0.23.20" with ring patch
+tokio-rustls = { git = "https://github.com/GreptimeTeam/tokio-rustls", rev = "4604ca6" } # version = "0.26.0" with ring patch
 # This is commented, since we are not using aws-lc-sys, if we need to use it, we need to uncomment this line or use a release after this commit, or it wouldn't compile with gcc < 8.1
 # see https://github.com/aws/aws-lc-rs/pull/526
 # aws-lc-sys = { git ="https://github.com/aws/aws-lc-rs", rev = "556558441e3494af4b156ae95ebc07ebc2fd38aa" }

Makefile

@@ -8,7 +8,7 @@ CARGO_BUILD_OPTS := --locked
 IMAGE_REGISTRY ?= docker.io
 IMAGE_NAMESPACE ?= greptime
 IMAGE_TAG ?= latest
-DEV_BUILDER_IMAGE_TAG ?= 2024-10-19-a5c00e85-20241024184445
+DEV_BUILDER_IMAGE_TAG ?= 2024-12-25-9d0fa5d5-20250124085746
 BUILDX_MULTI_PLATFORM_BUILD ?= false
 BUILDX_BUILDER_NAME ?= gtbuilder
 BASE_IMAGE ?= ubuntu
@@ -165,15 +165,14 @@ nextest: ## Install nextest tools.
 sqlness-test: ## Run sqlness test.
 	cargo sqlness ${SQLNESS_OPTS}
 
-# Run fuzz test ${FUZZ_TARGET}.
 RUNS ?= 1
 FUZZ_TARGET ?= fuzz_alter_table
 .PHONY: fuzz
-fuzz:
+fuzz: ## Run fuzz test ${FUZZ_TARGET}.
 	cargo fuzz run ${FUZZ_TARGET} --fuzz-dir tests-fuzz -D -s none -- -runs=${RUNS}
 
 .PHONY: fuzz-ls
-fuzz-ls:
+fuzz-ls: ## List all fuzz targets.
 	cargo fuzz list --fuzz-dir tests-fuzz
 
 .PHONY: check

README.md

@@ -138,7 +138,8 @@ Check the prerequisite:
 
 * [Rust toolchain](https://www.rust-lang.org/tools/install) (nightly)
 * [Protobuf compiler](https://grpc.io/docs/protoc-installation/) (>= 3.15)
-* Python toolchain (optional): Required only if built with PyO3 backend. More details for compiling with PyO3 can be found in its [documentation](https://pyo3.rs/v0.18.1/building_and_distribution#configuring-the-python-version).
+* C/C++ building essentials, including `gcc`/`g++`/`autoconf` and glibc library (eg. `libc6-dev` on Ubuntu and `glibc-devel` on Fedora)
+* Python toolchain (optional): Required only if using some test scripts.
 
 Build GreptimeDB binary:
 
@@ -228,4 +229,3 @@ Special thanks to all the contributors who have propelled GreptimeDB forward. Fo
 - GreptimeDB's query engine is powered by [Apache Arrow DataFusion™](https://arrow.apache.org/datafusion/).
 - [Apache OpenDAL™](https://opendal.apache.org) gives GreptimeDB a very general and elegant data access abstraction layer.
 - GreptimeDB's meta service is based on [etcd](https://etcd.io/).
-- GreptimeDB uses [RustPython](https://github.com/RustPython/RustPython) for experimental embedded python scripting.

config/config.md

@@ -18,6 +18,7 @@
 | `init_regions_parallelism` | Integer | `16` | Parallelism of initializing regions. |
 | `max_concurrent_queries` | Integer | `0` | The maximum current queries allowed to be executed. Zero means unlimited. |
 | `enable_telemetry` | Bool | `true` | Enable telemetry to collect anonymous usage data. Enabled by default. |
+| `max_in_flight_write_bytes` | String | Unset | The maximum in-flight write bytes. |
 | `runtime` | -- | -- | The runtime options. |
 | `runtime.global_rt_size` | Integer | `8` | The number of threads to execute the runtime for global read operations. |
 | `runtime.compact_rt_size` | Integer | `4` | The number of threads to execute the runtime for global write operations. |
@@ -25,6 +26,8 @@
 | `http.addr` | String | `127.0.0.1:4000` | The address to bind the HTTP server. |
 | `http.timeout` | String | `30s` | HTTP request timeout. Set to 0 to disable timeout. |
 | `http.body_limit` | String | `64MB` | HTTP request body limit.<br/>The following units are supported: `B`, `KB`, `KiB`, `MB`, `MiB`, `GB`, `GiB`, `TB`, `TiB`, `PB`, `PiB`.<br/>Set to 0 to disable limit. |
+| `http.enable_cors` | Bool | `true` | HTTP CORS support, it's turned on by default<br/>This allows browser to access http APIs without CORS restrictions |
+| `http.cors_allowed_origins` | Array | Unset | Customize allowed origins for HTTP CORS. |
 | `grpc` | -- | -- | The gRPC server options. |
 | `grpc.addr` | String | `127.0.0.1:4001` | The address to bind the gRPC server. |
 | `grpc.runtime_size` | Integer | `8` | The number of server worker threads. |
@@ -90,10 +93,12 @@
 | `procedure` | -- | -- | Procedure storage options. |
 | `procedure.max_retry_times` | Integer | `3` | Procedure max retry time. |
 | `procedure.retry_delay` | String | `500ms` | Initial retry delay of procedures, increases exponentially |
+| `flow` | -- | -- | flow engine options. |
+| `flow.num_workers` | Integer | `0` | The number of flow worker in flownode.<br/>Not setting(or set to 0) this value will use the number of CPU cores divided by 2. |
 | `storage` | -- | -- | The data storage options. |
 | `storage.data_home` | String | `/tmp/greptimedb/` | The working home directory. |
 | `storage.type` | String | `File` | The storage type used to store the data.<br/>- `File`: the data is stored in the local file system.<br/>- `S3`: the data is stored in the S3 object storage.<br/>- `Gcs`: the data is stored in the Google Cloud Storage.<br/>- `Azblob`: the data is stored in the Azure Blob Storage.<br/>- `Oss`: the data is stored in the Aliyun OSS. |
-| `storage.cache_path` | String | Unset | Read cache configuration for object storage such as 'S3' etc, it's configured by default when using object storage. It is recommended to configure it when using object storage for better performance.<br/>A local file directory, defaults to `{data_home}/object_cache/read`. An empty string means disabling. |
+| `storage.cache_path` | String | Unset | Read cache configuration for object storage such as 'S3' etc, it's configured by default when using object storage. It is recommended to configure it when using object storage for better performance.<br/>A local file directory, defaults to `{data_home}`. An empty string means disabling. |
 | `storage.cache_capacity` | String | Unset | The local file cache capacity in bytes. If your disk space is sufficient, it is recommended to set it larger. |
 | `storage.bucket` | String | Unset | The S3 bucket name.<br/>**It's only used when the storage type is `S3`, `Oss` and `Gcs`**. |
 | `storage.root` | String | Unset | The S3 data will be stored in the specified prefix, for example, `s3://${bucket}/${root}`.<br/>**It's only used when the storage type is `S3`, `Oss` and `Azblob`**. |
@@ -131,10 +136,10 @@
 | `region_engine.mito.vector_cache_size` | String | Auto | Cache size for vectors and arrow arrays. Setting it to 0 to disable the cache.<br/>If not set, it's default to 1/16 of OS memory with a max limitation of 512MB. |
 | `region_engine.mito.page_cache_size` | String | Auto | Cache size for pages of SST row groups. Setting it to 0 to disable the cache.<br/>If not set, it's default to 1/8 of OS memory. |
 | `region_engine.mito.selector_result_cache_size` | String | Auto | Cache size for time series selector (e.g. `last_value()`). Setting it to 0 to disable the cache.<br/>If not set, it's default to 1/16 of OS memory with a max limitation of 512MB. |
-| `region_engine.mito.enable_experimental_write_cache` | Bool | `false` | Whether to enable the experimental write cache, it's enabled by default when using object storage. It is recommended to enable it when using object storage for better performance. |
-| `region_engine.mito.experimental_write_cache_path` | String | `""` | File system path for write cache, defaults to `{data_home}/object_cache/write`. |
-| `region_engine.mito.experimental_write_cache_size` | String | `5GiB` | Capacity for write cache. If your disk space is sufficient, it is recommended to set it larger. |
-| `region_engine.mito.experimental_write_cache_ttl` | String | Unset | TTL for write cache. |
+| `region_engine.mito.enable_write_cache` | Bool | `false` | Whether to enable the write cache, it's enabled by default when using object storage. It is recommended to enable it when using object storage for better performance. |
+| `region_engine.mito.write_cache_path` | String | `""` | File system path for write cache, defaults to `{data_home}`. |
+| `region_engine.mito.write_cache_size` | String | `5GiB` | Capacity for write cache. If your disk space is sufficient, it is recommended to set it larger. |
+| `region_engine.mito.write_cache_ttl` | String | Unset | TTL for write cache. |
 | `region_engine.mito.sst_write_buffer_size` | String | `8MB` | Buffer size for SST writing. |
 | `region_engine.mito.parallel_scan_channel_size` | Integer | `32` | Capacity of the channel to send data from parallel scan tasks to the main task. |
 | `region_engine.mito.allow_stale_entries` | Bool | `false` | Whether to allow stale WAL entries read during replay. |
@@ -142,26 +147,33 @@
 | `region_engine.mito.index` | -- | -- | The options for index in Mito engine. |
 | `region_engine.mito.index.aux_path` | String | `""` | Auxiliary directory path for the index in filesystem, used to store intermediate files for<br/>creating the index and staging files for searching the index, defaults to `{data_home}/index_intermediate`.<br/>The default name for this directory is `index_intermediate` for backward compatibility.<br/><br/>This path contains two subdirectories:<br/>- `__intm`: for storing intermediate files used during creating index.<br/>- `staging`: for storing staging files used during searching index. |
 | `region_engine.mito.index.staging_size` | String | `2GB` | The max capacity of the staging directory. |
+| `region_engine.mito.index.metadata_cache_size` | String | `64MiB` | Cache size for inverted index metadata. |
+| `region_engine.mito.index.content_cache_size` | String | `128MiB` | Cache size for inverted index content. |
+| `region_engine.mito.index.content_cache_page_size` | String | `64KiB` | Page size for inverted index content cache. |
 | `region_engine.mito.inverted_index` | -- | -- | The options for inverted index in Mito engine. |
 | `region_engine.mito.inverted_index.create_on_flush` | String | `auto` | Whether to create the index on flush.<br/>- `auto`: automatically (default)<br/>- `disable`: never |
 | `region_engine.mito.inverted_index.create_on_compaction` | String | `auto` | Whether to create the index on compaction.<br/>- `auto`: automatically (default)<br/>- `disable`: never |
 | `region_engine.mito.inverted_index.apply_on_query` | String | `auto` | Whether to apply the index on query<br/>- `auto`: automatically (default)<br/>- `disable`: never |
 | `region_engine.mito.inverted_index.mem_threshold_on_create` | String | `auto` | Memory threshold for performing an external sort during index creation.<br/>- `auto`: automatically determine the threshold based on the system memory size (default)<br/>- `unlimited`: no memory limit<br/>- `[size]` e.g. `64MB`: fixed memory threshold |
 | `region_engine.mito.inverted_index.intermediate_path` | String | `""` | Deprecated, use `region_engine.mito.index.aux_path` instead. |
-| `region_engine.mito.inverted_index.metadata_cache_size` | String | `64MiB` | Cache size for inverted index metadata. |
-| `region_engine.mito.inverted_index.content_cache_size` | String | `128MiB` | Cache size for inverted index content. |
-| `region_engine.mito.inverted_index.content_cache_page_size` | String | `8MiB` | Page size for inverted index content cache. |
 | `region_engine.mito.fulltext_index` | -- | -- | The options for full-text index in Mito engine. |
 | `region_engine.mito.fulltext_index.create_on_flush` | String | `auto` | Whether to create the index on flush.<br/>- `auto`: automatically (default)<br/>- `disable`: never |
 | `region_engine.mito.fulltext_index.create_on_compaction` | String | `auto` | Whether to create the index on compaction.<br/>- `auto`: automatically (default)<br/>- `disable`: never |
 | `region_engine.mito.fulltext_index.apply_on_query` | String | `auto` | Whether to apply the index on query<br/>- `auto`: automatically (default)<br/>- `disable`: never |
 | `region_engine.mito.fulltext_index.mem_threshold_on_create` | String | `auto` | Memory threshold for index creation.<br/>- `auto`: automatically determine the threshold based on the system memory size (default)<br/>- `unlimited`: no memory limit<br/>- `[size]` e.g. `64MB`: fixed memory threshold |
+| `region_engine.mito.bloom_filter_index` | -- | -- | The options for bloom filter in Mito engine. |
+| `region_engine.mito.bloom_filter_index.create_on_flush` | String | `auto` | Whether to create the bloom filter on flush.<br/>- `auto`: automatically (default)<br/>- `disable`: never |
+| `region_engine.mito.bloom_filter_index.create_on_compaction` | String | `auto` | Whether to create the bloom filter on compaction.<br/>- `auto`: automatically (default)<br/>- `disable`: never |
+| `region_engine.mito.bloom_filter_index.apply_on_query` | String | `auto` | Whether to apply the bloom filter on query<br/>- `auto`: automatically (default)<br/>- `disable`: never |
+| `region_engine.mito.bloom_filter_index.mem_threshold_on_create` | String | `auto` | Memory threshold for bloom filter creation.<br/>- `auto`: automatically determine the threshold based on the system memory size (default)<br/>- `unlimited`: no memory limit<br/>- `[size]` e.g. `64MB`: fixed memory threshold |
 | `region_engine.mito.memtable` | -- | -- | -- |
 | `region_engine.mito.memtable.type` | String | `time_series` | Memtable type.<br/>- `time_series`: time-series memtable<br/>- `partition_tree`: partition tree memtable (experimental) |
 | `region_engine.mito.memtable.index_max_keys_per_shard` | Integer | `8192` | The max number of keys in one shard.<br/>Only available for `partition_tree` memtable. |
 | `region_engine.mito.memtable.data_freeze_threshold` | Integer | `32768` | The max rows of data inside the actively writing buffer in one shard.<br/>Only available for `partition_tree` memtable. |
 | `region_engine.mito.memtable.fork_dictionary_bytes` | String | `1GiB` | Max dictionary bytes.<br/>Only available for `partition_tree` memtable. |
 | `region_engine.file` | -- | -- | Enable the file engine. |
+| `region_engine.metric` | -- | -- | Metric engine options. |
+| `region_engine.metric.experimental_sparse_primary_key_encoding` | Bool | `false` | Whether to enable the experimental sparse primary key encoding. |
 | `logging` | -- | -- | The logging options. |
 | `logging.dir` | String | `/tmp/greptimedb/logs` | The directory to store the log files. If set to empty, logs will not be written to files. |
 | `logging.level` | String | Unset | The log level. Can be `info`/`debug`/`warn`/`error`. |
@@ -195,6 +207,7 @@
 | Key | Type | Default | Descriptions |
 | --- | -----| ------- | ----------- |
 | `default_timezone` | String | Unset | The default timezone of the server. |
+| `max_in_flight_write_bytes` | String | Unset | The maximum in-flight write bytes. |
 | `runtime` | -- | -- | The runtime options. |
 | `runtime.global_rt_size` | Integer | `8` | The number of threads to execute the runtime for global read operations. |
 | `runtime.compact_rt_size` | Integer | `4` | The number of threads to execute the runtime for global write operations. |
@@ -205,9 +218,11 @@
 | `http.addr` | String | `127.0.0.1:4000` | The address to bind the HTTP server. |
 | `http.timeout` | String | `30s` | HTTP request timeout. Set to 0 to disable timeout. |
 | `http.body_limit` | String | `64MB` | HTTP request body limit.<br/>The following units are supported: `B`, `KB`, `KiB`, `MB`, `MiB`, `GB`, `GiB`, `TB`, `TiB`, `PB`, `PiB`.<br/>Set to 0 to disable limit. |
+| `http.enable_cors` | Bool | `true` | HTTP CORS support, it's turned on by default<br/>This allows browser to access http APIs without CORS restrictions |
+| `http.cors_allowed_origins` | Array | Unset | Customize allowed origins for HTTP CORS. |
 | `grpc` | -- | -- | The gRPC server options. |
 | `grpc.addr` | String | `127.0.0.1:4001` | The address to bind the gRPC server. |
-| `grpc.hostname` | String | `127.0.0.1` | The hostname advertised to the metasrv,<br/>and used for connections from outside the host |
+| `grpc.hostname` | String | `127.0.0.1:4001` | The hostname advertised to the metasrv,<br/>and used for connections from outside the host |
 | `grpc.runtime_size` | Integer | `8` | The number of server worker threads. |
 | `grpc.tls` | -- | -- | gRPC server TLS options, see `mysql.tls` section. |
 | `grpc.tls.mode` | String | `disable` | TLS mode. |
@@ -286,9 +301,11 @@
 | `data_home` | String | `/tmp/metasrv/` | The working home directory. |
 | `bind_addr` | String | `127.0.0.1:3002` | The bind address of metasrv. |
 | `server_addr` | String | `127.0.0.1:3002` | The communication server address for frontend and datanode to connect to metasrv,  "127.0.0.1:3002" by default for localhost. |
-| `store_addrs` | Array | -- | Store server address default to etcd store. |
+| `store_addrs` | Array | -- | Store server address default to etcd store.<br/>For postgres store, the format is:<br/>"password=password dbname=postgres user=postgres host=localhost port=5432"<br/>For etcd store, the format is:<br/>"127.0.0.1:2379" |
 | `store_key_prefix` | String | `""` | If it's not empty, the metasrv will store all data with this key prefix. |
-| `backend` | String | `EtcdStore` | The datastore for meta server. |
+| `backend` | String | `etcd_store` | The datastore for meta server.<br/>Available values:<br/>- `etcd_store` (default value)<br/>- `memory_store`<br/>- `postgres_store` |
+| `meta_table_name` | String | `greptime_metakv` | Table name in RDS to store metadata. Effect when using a RDS kvbackend.<br/>**Only used when backend is `postgres_store`.** |
+| `meta_election_lock_id` | Integer | `1` | Advisory lock id in PostgreSQL for election. Effect when using PostgreSQL as kvbackend<br/>Only used when backend is `postgres_store`. |
 | `selector` | String | `round_robin` | Datanode selector type.<br/>- `round_robin` (default value)<br/>- `lease_based`<br/>- `load_based`<br/>For details, please see "https://docs.greptime.com/developer-guide/metasrv/selector". |
 | `use_memory_store` | Bool | `false` | Store data in memory. |
 | `enable_region_failover` | Bool | `false` | Whether to enable region failover.<br/>This feature is only available on GreptimeDB running on cluster mode and<br/>- Using Remote WAL<br/>- Using shared storage (e.g., s3). |
@@ -316,7 +333,7 @@
 | `wal.auto_create_topics` | Bool | `true` | Automatically create topics for WAL.<br/>Set to `true` to automatically create topics for WAL.<br/>Otherwise, use topics named `topic_name_prefix_[0..num_topics)` |
 | `wal.num_topics` | Integer | `64` | Number of topics. |
 | `wal.selector_type` | String | `round_robin` | Topic selector type.<br/>Available selector types:<br/>- `round_robin` (default) |
-| `wal.topic_name_prefix` | String | `greptimedb_wal_topic` | A Kafka topic is constructed by concatenating `topic_name_prefix` and `topic_id`.<br/>i.g., greptimedb_wal_topic_0, greptimedb_wal_topic_1. |
+| `wal.topic_name_prefix` | String | `greptimedb_wal_topic` | A Kafka topic is constructed by concatenating `topic_name_prefix` and `topic_id`.<br/>Only accepts strings that match the following regular expression pattern:<br/>[a-zA-Z_:-][a-zA-Z0-9_:\-\.@#]*<br/>i.g., greptimedb_wal_topic_0, greptimedb_wal_topic_1. |
 | `wal.replication_factor` | Integer | `1` | Expected number of replicas of each partition. |
 | `wal.create_topic_timeout` | String | `30s` | Above which a topic creation operation will be cancelled. |
 | `wal.backoff_init` | String | `500ms` | The initial backoff for kafka clients. |
@@ -371,7 +388,7 @@
 | `http.body_limit` | String | `64MB` | HTTP request body limit.<br/>The following units are supported: `B`, `KB`, `KiB`, `MB`, `MiB`, `GB`, `GiB`, `TB`, `TiB`, `PB`, `PiB`.<br/>Set to 0 to disable limit. |
 | `grpc` | -- | -- | The gRPC server options. |
 | `grpc.addr` | String | `127.0.0.1:3001` | The address to bind the gRPC server. |
-| `grpc.hostname` | String | `127.0.0.1` | The hostname advertised to the metasrv,<br/>and used for connections from outside the host |
+| `grpc.hostname` | String | `127.0.0.1:3001` | The hostname advertised to the metasrv,<br/>and used for connections from outside the host |
 | `grpc.runtime_size` | Integer | `8` | The number of server worker threads. |
 | `grpc.max_recv_message_size` | String | `512MB` | The maximum receive message size for gRPC server. |
 | `grpc.max_send_message_size` | String | `512MB` | The maximum send message size for gRPC server. |
@@ -421,7 +438,7 @@
 | `storage` | -- | -- | The data storage options. |
 | `storage.data_home` | String | `/tmp/greptimedb/` | The working home directory. |
 | `storage.type` | String | `File` | The storage type used to store the data.<br/>- `File`: the data is stored in the local file system.<br/>- `S3`: the data is stored in the S3 object storage.<br/>- `Gcs`: the data is stored in the Google Cloud Storage.<br/>- `Azblob`: the data is stored in the Azure Blob Storage.<br/>- `Oss`: the data is stored in the Aliyun OSS. |
-| `storage.cache_path` | String | Unset | Read cache configuration for object storage such as 'S3' etc, it's configured by default when using object storage. It is recommended to configure it when using object storage for better performance.<br/>A local file directory, defaults to `{data_home}/object_cache/read`. An empty string means disabling. |
+| `storage.cache_path` | String | Unset | Read cache configuration for object storage such as 'S3' etc, it's configured by default when using object storage. It is recommended to configure it when using object storage for better performance.<br/>A local file directory, defaults to `{data_home}`. An empty string means disabling. |
 | `storage.cache_capacity` | String | Unset | The local file cache capacity in bytes. If your disk space is sufficient, it is recommended to set it larger. |
 | `storage.bucket` | String | Unset | The S3 bucket name.<br/>**It's only used when the storage type is `S3`, `Oss` and `Gcs`**. |
 | `storage.root` | String | Unset | The S3 data will be stored in the specified prefix, for example, `s3://${bucket}/${root}`.<br/>**It's only used when the storage type is `S3`, `Oss` and `Azblob`**. |
@@ -459,10 +476,10 @@
 | `region_engine.mito.vector_cache_size` | String | Auto | Cache size for vectors and arrow arrays. Setting it to 0 to disable the cache.<br/>If not set, it's default to 1/16 of OS memory with a max limitation of 512MB. |
 | `region_engine.mito.page_cache_size` | String | Auto | Cache size for pages of SST row groups. Setting it to 0 to disable the cache.<br/>If not set, it's default to 1/8 of OS memory. |
 | `region_engine.mito.selector_result_cache_size` | String | Auto | Cache size for time series selector (e.g. `last_value()`). Setting it to 0 to disable the cache.<br/>If not set, it's default to 1/16 of OS memory with a max limitation of 512MB. |
-| `region_engine.mito.enable_experimental_write_cache` | Bool | `false` | Whether to enable the experimental write cache, it's enabled by default when using object storage. It is recommended to enable it when using object storage for better performance. |
-| `region_engine.mito.experimental_write_cache_path` | String | `""` | File system path for write cache, defaults to `{data_home}/object_cache/write`. |
-| `region_engine.mito.experimental_write_cache_size` | String | `5GiB` | Capacity for write cache. If your disk space is sufficient, it is recommended to set it larger. |
-| `region_engine.mito.experimental_write_cache_ttl` | String | Unset | TTL for write cache. |
+| `region_engine.mito.enable_write_cache` | Bool | `false` | Whether to enable the write cache, it's enabled by default when using object storage. It is recommended to enable it when using object storage for better performance. |
+| `region_engine.mito.write_cache_path` | String | `""` | File system path for write cache, defaults to `{data_home}`. |
+| `region_engine.mito.write_cache_size` | String | `5GiB` | Capacity for write cache. If your disk space is sufficient, it is recommended to set it larger. |
+| `region_engine.mito.write_cache_ttl` | String | Unset | TTL for write cache. |
 | `region_engine.mito.sst_write_buffer_size` | String | `8MB` | Buffer size for SST writing. |
 | `region_engine.mito.parallel_scan_channel_size` | Integer | `32` | Capacity of the channel to send data from parallel scan tasks to the main task. |
 | `region_engine.mito.allow_stale_entries` | Bool | `false` | Whether to allow stale WAL entries read during replay. |
@@ -470,26 +487,33 @@
 | `region_engine.mito.index` | -- | -- | The options for index in Mito engine. |
 | `region_engine.mito.index.aux_path` | String | `""` | Auxiliary directory path for the index in filesystem, used to store intermediate files for<br/>creating the index and staging files for searching the index, defaults to `{data_home}/index_intermediate`.<br/>The default name for this directory is `index_intermediate` for backward compatibility.<br/><br/>This path contains two subdirectories:<br/>- `__intm`: for storing intermediate files used during creating index.<br/>- `staging`: for storing staging files used during searching index. |
 | `region_engine.mito.index.staging_size` | String | `2GB` | The max capacity of the staging directory. |
+| `region_engine.mito.index.metadata_cache_size` | String | `64MiB` | Cache size for inverted index metadata. |
+| `region_engine.mito.index.content_cache_size` | String | `128MiB` | Cache size for inverted index content. |
+| `region_engine.mito.index.content_cache_page_size` | String | `64KiB` | Page size for inverted index content cache. |
 | `region_engine.mito.inverted_index` | -- | -- | The options for inverted index in Mito engine. |
 | `region_engine.mito.inverted_index.create_on_flush` | String | `auto` | Whether to create the index on flush.<br/>- `auto`: automatically (default)<br/>- `disable`: never |
 | `region_engine.mito.inverted_index.create_on_compaction` | String | `auto` | Whether to create the index on compaction.<br/>- `auto`: automatically (default)<br/>- `disable`: never |
 | `region_engine.mito.inverted_index.apply_on_query` | String | `auto` | Whether to apply the index on query<br/>- `auto`: automatically (default)<br/>- `disable`: never |
 | `region_engine.mito.inverted_index.mem_threshold_on_create` | String | `auto` | Memory threshold for performing an external sort during index creation.<br/>- `auto`: automatically determine the threshold based on the system memory size (default)<br/>- `unlimited`: no memory limit<br/>- `[size]` e.g. `64MB`: fixed memory threshold |
 | `region_engine.mito.inverted_index.intermediate_path` | String | `""` | Deprecated, use `region_engine.mito.index.aux_path` instead. |
-| `region_engine.mito.inverted_index.metadata_cache_size` | String | `64MiB` | Cache size for inverted index metadata. |
-| `region_engine.mito.inverted_index.content_cache_size` | String | `128MiB` | Cache size for inverted index content. |
-| `region_engine.mito.inverted_index.content_cache_page_size` | String | `8MiB` | Page size for inverted index content cache. |
 | `region_engine.mito.fulltext_index` | -- | -- | The options for full-text index in Mito engine. |
 | `region_engine.mito.fulltext_index.create_on_flush` | String | `auto` | Whether to create the index on flush.<br/>- `auto`: automatically (default)<br/>- `disable`: never |
 | `region_engine.mito.fulltext_index.create_on_compaction` | String | `auto` | Whether to create the index on compaction.<br/>- `auto`: automatically (default)<br/>- `disable`: never |
 | `region_engine.mito.fulltext_index.apply_on_query` | String | `auto` | Whether to apply the index on query<br/>- `auto`: automatically (default)<br/>- `disable`: never |
 | `region_engine.mito.fulltext_index.mem_threshold_on_create` | String | `auto` | Memory threshold for index creation.<br/>- `auto`: automatically determine the threshold based on the system memory size (default)<br/>- `unlimited`: no memory limit<br/>- `[size]` e.g. `64MB`: fixed memory threshold |
+| `region_engine.mito.bloom_filter_index` | -- | -- | The options for bloom filter index in Mito engine. |
+| `region_engine.mito.bloom_filter_index.create_on_flush` | String | `auto` | Whether to create the index on flush.<br/>- `auto`: automatically (default)<br/>- `disable`: never |
+| `region_engine.mito.bloom_filter_index.create_on_compaction` | String | `auto` | Whether to create the index on compaction.<br/>- `auto`: automatically (default)<br/>- `disable`: never |
+| `region_engine.mito.bloom_filter_index.apply_on_query` | String | `auto` | Whether to apply the index on query<br/>- `auto`: automatically (default)<br/>- `disable`: never |
+| `region_engine.mito.bloom_filter_index.mem_threshold_on_create` | String | `auto` | Memory threshold for the index creation.<br/>- `auto`: automatically determine the threshold based on the system memory size (default)<br/>- `unlimited`: no memory limit<br/>- `[size]` e.g. `64MB`: fixed memory threshold |
 | `region_engine.mito.memtable` | -- | -- | -- |
 | `region_engine.mito.memtable.type` | String | `time_series` | Memtable type.<br/>- `time_series`: time-series memtable<br/>- `partition_tree`: partition tree memtable (experimental) |
 | `region_engine.mito.memtable.index_max_keys_per_shard` | Integer | `8192` | The max number of keys in one shard.<br/>Only available for `partition_tree` memtable. |
 | `region_engine.mito.memtable.data_freeze_threshold` | Integer | `32768` | The max rows of data inside the actively writing buffer in one shard.<br/>Only available for `partition_tree` memtable. |
 | `region_engine.mito.memtable.fork_dictionary_bytes` | String | `1GiB` | Max dictionary bytes.<br/>Only available for `partition_tree` memtable. |
 | `region_engine.file` | -- | -- | Enable the file engine. |
+| `region_engine.metric` | -- | -- | Metric engine options. |
+| `region_engine.metric.experimental_sparse_primary_key_encoding` | Bool | `false` | Whether to enable the experimental sparse primary key encoding. |
 | `logging` | -- | -- | The logging options. |
 | `logging.dir` | String | `/tmp/greptimedb/logs` | The directory to store the log files. If set to empty, logs will not be written to files. |
 | `logging.level` | String | Unset | The log level. Can be `info`/`debug`/`warn`/`error`. |
@@ -522,12 +546,18 @@
 | --- | -----| ------- | ----------- |
 | `mode` | String | `distributed` | The running mode of the flownode. It can be `standalone` or `distributed`. |
 | `node_id` | Integer | Unset | The flownode identifier and should be unique in the cluster. |
+| `flow` | -- | -- | flow engine options. |
+| `flow.num_workers` | Integer | `0` | The number of flow worker in flownode.<br/>Not setting(or set to 0) this value will use the number of CPU cores divided by 2. |
 | `grpc` | -- | -- | The gRPC server options. |
 | `grpc.addr` | String | `127.0.0.1:6800` | The address to bind the gRPC server. |
 | `grpc.hostname` | String | `127.0.0.1` | The hostname advertised to the metasrv,<br/>and used for connections from outside the host |
 | `grpc.runtime_size` | Integer | `2` | The number of server worker threads. |
 | `grpc.max_recv_message_size` | String | `512MB` | The maximum receive message size for gRPC server. |
 | `grpc.max_send_message_size` | String | `512MB` | The maximum send message size for gRPC server. |
+| `http` | -- | -- | The HTTP server options. |
+| `http.addr` | String | `127.0.0.1:4000` | The address to bind the HTTP server. |
+| `http.timeout` | String | `30s` | HTTP request timeout. Set to 0 to disable timeout. |
+| `http.body_limit` | String | `64MB` | HTTP request body limit.<br/>The following units are supported: `B`, `KB`, `KiB`, `MB`, `MiB`, `GB`, `GiB`, `TB`, `TiB`, `PB`, `PiB`.<br/>Set to 0 to disable limit. |
 | `meta_client` | -- | -- | The metasrv client options. |
 | `meta_client.metasrv_addrs` | Array | -- | The addresses of the metasrv. |
 | `meta_client.timeout` | String | `3s` | Operation timeout. |

config/datanode.example.toml

@@ -59,7 +59,7 @@ body_limit = "64MB"
 addr = "127.0.0.1:3001"
 ## The hostname advertised to the metasrv,
 ## and used for connections from outside the host
-hostname = "127.0.0.1"
+hostname = "127.0.0.1:3001"
 ## The number of server worker threads.
 runtime_size = 8
 ## The maximum receive message size for gRPC server.
@@ -294,7 +294,7 @@ data_home = "/tmp/greptimedb/"
 type = "File"
 
 ## Read cache configuration for object storage such as 'S3' etc, it's configured by default when using object storage. It is recommended to configure it when using object storage for better performance.
-## A local file directory, defaults to `{data_home}/object_cache/read`. An empty string means disabling.
+## A local file directory, defaults to `{data_home}`. An empty string means disabling.
 ## @toml2docs:none-default
 #+ cache_path = ""
 
@@ -475,18 +475,18 @@ auto_flush_interval = "1h"
 ## @toml2docs:none-default="Auto"
 #+ selector_result_cache_size = "512MB"
 
-## Whether to enable the experimental write cache, it's enabled by default when using object storage. It is recommended to enable it when using object storage for better performance.
-enable_experimental_write_cache = false
+## Whether to enable the write cache, it's enabled by default when using object storage. It is recommended to enable it when using object storage for better performance.
+enable_write_cache = false
 
-## File system path for write cache, defaults to `{data_home}/object_cache/write`.
-experimental_write_cache_path = ""
+## File system path for write cache, defaults to `{data_home}`.
+write_cache_path = ""
 
 ## Capacity for write cache. If your disk space is sufficient, it is recommended to set it larger.
-experimental_write_cache_size = "5GiB"
+write_cache_size = "5GiB"
 
 ## TTL for write cache.
 ## @toml2docs:none-default
-experimental_write_cache_ttl = "8h"
+write_cache_ttl = "8h"
 
 ## Buffer size for SST writing.
 sst_write_buffer_size = "8MB"
@@ -516,6 +516,15 @@ aux_path = ""
 ## The max capacity of the staging directory.
 staging_size = "2GB"
 
+## Cache size for inverted index metadata.
+metadata_cache_size = "64MiB"
+
+## Cache size for inverted index content.
+content_cache_size = "128MiB"
+
+## Page size for inverted index content cache.
+content_cache_page_size = "64KiB"
+
 ## The options for inverted index in Mito engine.
 [region_engine.mito.inverted_index]
 
@@ -543,15 +552,6 @@ mem_threshold_on_create = "auto"
 ## Deprecated, use `region_engine.mito.index.aux_path` instead.
 intermediate_path = ""
 
-## Cache size for inverted index metadata.
-metadata_cache_size = "64MiB"
-
-## Cache size for inverted index content.
-content_cache_size = "128MiB"
-
-## Page size for inverted index content cache.
-content_cache_page_size = "8MiB"
-
 ## The options for full-text index in Mito engine.
 [region_engine.mito.fulltext_index]
 
@@ -576,6 +576,30 @@ apply_on_query = "auto"
 ## - `[size]` e.g. `64MB`: fixed memory threshold
 mem_threshold_on_create = "auto"
 
+## The options for bloom filter index in Mito engine.
+[region_engine.mito.bloom_filter_index]
+
+## Whether to create the index on flush.
+## - `auto`: automatically (default)
+## - `disable`: never
+create_on_flush = "auto"
+
+## Whether to create the index on compaction.
+## - `auto`: automatically (default)
+## - `disable`: never
+create_on_compaction = "auto"
+
+## Whether to apply the index on query
+## - `auto`: automatically (default)
+## - `disable`: never
+apply_on_query = "auto"
+
+## Memory threshold for the index creation.
+## - `auto`: automatically determine the threshold based on the system memory size (default)
+## - `unlimited`: no memory limit
+## - `[size]` e.g. `64MB`: fixed memory threshold
+mem_threshold_on_create = "auto"
+
 [region_engine.mito.memtable]
 ## Memtable type.
 ## - `time_series`: time-series memtable
@@ -598,6 +622,12 @@ fork_dictionary_bytes = "1GiB"
 ## Enable the file engine.
 [region_engine.file]
 
+[[region_engine]]
+## Metric engine options.
+[region_engine.metric]
+## Whether to enable the experimental sparse primary key encoding.
+experimental_sparse_primary_key_encoding = false
+
 ## The logging options.
 [logging]
 ## The directory to store the log files. If set to empty, logs will not be written to files.

config/flownode.example.toml

@@ -5,6 +5,12 @@ mode = "distributed"
 ## @toml2docs:none-default
 node_id = 14
 
+## flow engine options.
+[flow]
+## The number of flow worker in flownode.
+## Not setting(or set to 0) this value will use the number of CPU cores divided by 2.
+#+num_workers=0
+
 ## The gRPC server options.
 [grpc]
 ## The address to bind the gRPC server.
@@ -19,6 +25,16 @@ max_recv_message_size = "512MB"
 ## The maximum send message size for gRPC server.
 max_send_message_size = "512MB"
 
+## The HTTP server options.
+[http]
+## The address to bind the HTTP server.
+addr = "127.0.0.1:4000"
+## HTTP request timeout. Set to 0 to disable timeout.
+timeout = "30s"
+## HTTP request body limit.
+## The following units are supported: `B`, `KB`, `KiB`, `MB`, `MiB`, `GB`, `GiB`, `TB`, `TiB`, `PB`, `PiB`.
+## Set to 0 to disable limit.
+body_limit = "64MB"
 
 ## The metasrv client options.
 [meta_client]

config/frontend.example.toml

@@ -2,6 +2,10 @@
 ## @toml2docs:none-default
 default_timezone = "UTC"
 
+## The maximum in-flight write bytes.
+## @toml2docs:none-default
+#+ max_in_flight_write_bytes = "500MB"
+
 ## The runtime options.
 #+ [runtime]
 ## The number of threads to execute the runtime for global read operations.
@@ -27,6 +31,12 @@ timeout = "30s"
 ## The following units are supported: `B`, `KB`, `KiB`, `MB`, `MiB`, `GB`, `GiB`, `TB`, `TiB`, `PB`, `PiB`.
 ## Set to 0 to disable limit.
 body_limit = "64MB"
+## HTTP CORS support, it's turned on by default
+## This allows browser to access http APIs without CORS restrictions
+enable_cors = true
+## Customize allowed origins for HTTP CORS.
+## @toml2docs:none-default
+cors_allowed_origins = ["https://example.com"]
 
 ## The gRPC server options.
 [grpc]
@@ -34,7 +44,7 @@ body_limit = "64MB"
 addr = "127.0.0.1:4001"
 ## The hostname advertised to the metasrv,
 ## and used for connections from outside the host
-hostname = "127.0.0.1"
+hostname = "127.0.0.1:4001"
 ## The number of server worker threads.
 runtime_size = 8
 

config/metasrv.example.toml

@@ -8,13 +8,29 @@ bind_addr = "127.0.0.1:3002"
 server_addr = "127.0.0.1:3002"
 
 ## Store server address default to etcd store.
+## For postgres store, the format is:
+## "password=password dbname=postgres user=postgres host=localhost port=5432"
+## For etcd store, the format is:
+## "127.0.0.1:2379"
 store_addrs = ["127.0.0.1:2379"]
 
 ## If it's not empty, the metasrv will store all data with this key prefix.
 store_key_prefix = ""
 
 ## The datastore for meta server.
-backend = "EtcdStore"
+## Available values:
+## - `etcd_store` (default value)
+## - `memory_store`
+## - `postgres_store`
+backend = "etcd_store"
+
+## Table name in RDS to store metadata. Effect when using a RDS kvbackend.
+## **Only used when backend is `postgres_store`.**
+meta_table_name = "greptime_metakv"
+
+## Advisory lock id in PostgreSQL for election. Effect when using PostgreSQL as kvbackend
+## Only used when backend is `postgres_store`.
+meta_election_lock_id = 1
 
 ## Datanode selector type.
 ## - `round_robin` (default value)
@@ -113,6 +129,8 @@ num_topics = 64
 selector_type = "round_robin"
 
 ## A Kafka topic is constructed by concatenating `topic_name_prefix` and `topic_id`.
+## Only accepts strings that match the following regular expression pattern:
+## [a-zA-Z_:-][a-zA-Z0-9_:\-\.@#]*
 ## i.g., greptimedb_wal_topic_0, greptimedb_wal_topic_1.
 topic_name_prefix = "greptimedb_wal_topic"
 

config/standalone.example.toml

@@ -18,6 +18,10 @@ max_concurrent_queries = 0
 ## Enable telemetry to collect anonymous usage data. Enabled by default.
 #+ enable_telemetry = true
 
+## The maximum in-flight write bytes.
+## @toml2docs:none-default
+#+ max_in_flight_write_bytes = "500MB"
+
 ## The runtime options.
 #+ [runtime]
 ## The number of threads to execute the runtime for global read operations.
@@ -35,6 +39,12 @@ timeout = "30s"
 ## The following units are supported: `B`, `KB`, `KiB`, `MB`, `MiB`, `GB`, `GiB`, `TB`, `TiB`, `PB`, `PiB`.
 ## Set to 0 to disable limit.
 body_limit = "64MB"
+## HTTP CORS support, it's turned on by default
+## This allows browser to access http APIs without CORS restrictions
+enable_cors = true
+## Customize allowed origins for HTTP CORS.
+## @toml2docs:none-default
+cors_allowed_origins = ["https://example.com"]
 
 ## The gRPC server options.
 [grpc]
@@ -280,6 +290,12 @@ max_retry_times = 3
 ## Initial retry delay of procedures, increases exponentially
 retry_delay = "500ms"
 
+## flow engine options.
+[flow]
+## The number of flow worker in flownode.
+## Not setting(or set to 0) this value will use the number of CPU cores divided by 2.
+#+num_workers=0
+
 # Example of using S3 as the storage.
 # [storage]
 # type = "S3"
@@ -333,7 +349,7 @@ data_home = "/tmp/greptimedb/"
 type = "File"
 
 ## Read cache configuration for object storage such as 'S3' etc, it's configured by default when using object storage. It is recommended to configure it when using object storage for better performance.
-## A local file directory, defaults to `{data_home}/object_cache/read`. An empty string means disabling.
+## A local file directory, defaults to `{data_home}`. An empty string means disabling.
 ## @toml2docs:none-default
 #+ cache_path = ""
 
@@ -514,18 +530,18 @@ auto_flush_interval = "1h"
 ## @toml2docs:none-default="Auto"
 #+ selector_result_cache_size = "512MB"
 
-## Whether to enable the experimental write cache, it's enabled by default when using object storage. It is recommended to enable it when using object storage for better performance.
-enable_experimental_write_cache = false
+## Whether to enable the write cache, it's enabled by default when using object storage. It is recommended to enable it when using object storage for better performance.
+enable_write_cache = false
 
-## File system path for write cache, defaults to `{data_home}/object_cache/write`.
-experimental_write_cache_path = ""
+## File system path for write cache, defaults to `{data_home}`.
+write_cache_path = ""
 
 ## Capacity for write cache. If your disk space is sufficient, it is recommended to set it larger.
-experimental_write_cache_size = "5GiB"
+write_cache_size = "5GiB"
 
 ## TTL for write cache.
 ## @toml2docs:none-default
-experimental_write_cache_ttl = "8h"
+write_cache_ttl = "8h"
 
 ## Buffer size for SST writing.
 sst_write_buffer_size = "8MB"
@@ -555,6 +571,15 @@ aux_path = ""
 ## The max capacity of the staging directory.
 staging_size = "2GB"
 
+## Cache size for inverted index metadata.
+metadata_cache_size = "64MiB"
+
+## Cache size for inverted index content.
+content_cache_size = "128MiB"
+
+## Page size for inverted index content cache.
+content_cache_page_size = "64KiB"
+
 ## The options for inverted index in Mito engine.
 [region_engine.mito.inverted_index]
 
@@ -582,15 +607,6 @@ mem_threshold_on_create = "auto"
 ## Deprecated, use `region_engine.mito.index.aux_path` instead.
 intermediate_path = ""
 
-## Cache size for inverted index metadata.
-metadata_cache_size = "64MiB"
-
-## Cache size for inverted index content.
-content_cache_size = "128MiB"
-
-## Page size for inverted index content cache.
-content_cache_page_size = "8MiB"
-
 ## The options for full-text index in Mito engine.
 [region_engine.mito.fulltext_index]
 
@@ -615,6 +631,30 @@ apply_on_query = "auto"
 ## - `[size]` e.g. `64MB`: fixed memory threshold
 mem_threshold_on_create = "auto"
 
+## The options for bloom filter in Mito engine.
+[region_engine.mito.bloom_filter_index]
+
+## Whether to create the bloom filter on flush.
+## - `auto`: automatically (default)
+## - `disable`: never
+create_on_flush = "auto"
+
+## Whether to create the bloom filter on compaction.
+## - `auto`: automatically (default)
+## - `disable`: never
+create_on_compaction = "auto"
+
+## Whether to apply the bloom filter on query
+## - `auto`: automatically (default)
+## - `disable`: never
+apply_on_query = "auto"
+
+## Memory threshold for bloom filter creation.
+## - `auto`: automatically determine the threshold based on the system memory size (default)
+## - `unlimited`: no memory limit
+## - `[size]` e.g. `64MB`: fixed memory threshold
+mem_threshold_on_create = "auto"
+
 [region_engine.mito.memtable]
 ## Memtable type.
 ## - `time_series`: time-series memtable
@@ -637,6 +677,12 @@ fork_dictionary_bytes = "1GiB"
 ## Enable the file engine.
 [region_engine.file]
 
+[[region_engine]]
+## Metric engine options.
+[region_engine.metric]
+## Whether to enable the experimental sparse primary key encoding.
+experimental_sparse_primary_key_encoding = false
+
 ## The logging options.
 [logging]
 ## The directory to store the log files. If set to empty, logs will not be written to files.

cyborg/bin/bump-doc-version.ts

@@ -0,0 +1,75 @@
+/*
+ * Copyright 2023 Greptime Team
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ *     http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+import * as core from "@actions/core";
+import {obtainClient} from "@/common";
+
+async function triggerWorkflow(workflowId: string, version: string) {
+  const docsClient = obtainClient("DOCS_REPO_TOKEN")
+  try {
+    await docsClient.rest.actions.createWorkflowDispatch({
+      owner: "GreptimeTeam",
+      repo: "docs",
+      workflow_id: workflowId,
+      ref: "main",
+      inputs: {
+        version,
+      },
+    });
+    console.log(`Successfully triggered ${workflowId} workflow with version ${version}`);
+  } catch (error) {
+    core.setFailed(`Failed to trigger workflow: ${error.message}`);
+  }
+}
+
+function determineWorkflow(version: string): [string, string] {
+  // Check if it's a nightly version
+  if (version.includes('nightly')) {
+    return ['bump-nightly-version.yml', version];
+  }
+
+  const parts = version.split('.');
+
+  if (parts.length !== 3) {
+    throw new Error('Invalid version format');
+  }
+
+  // If patch version (last number) is 0, it's a major version
+  // Return only major.minor version
+  if (parts[2] === '0') {
+    return ['bump-version.yml', `${parts[0]}.${parts[1]}`];
+  }
+
+  // Otherwise it's a patch version, use full version
+  return ['bump-patch-version.yml', version];
+}
+
+const version = process.env.VERSION;
+if (!version) {
+  core.setFailed("VERSION environment variable is required");
+  process.exit(1);
+}
+
+// Remove 'v' prefix if exists
+const cleanVersion = version.startsWith('v') ? version.slice(1) : version;
+
+try {
+  const [workflowId, apiVersion] = determineWorkflow(cleanVersion);
+  triggerWorkflow(workflowId, apiVersion);
+} catch (error) {
+  core.setFailed(`Error processing version: ${error.message}`);
+  process.exit(1);
+}

docker/buildx/centos/Dockerfile

@@ -13,8 +13,6 @@ RUN yum install -y epel-release  \
     openssl \
     openssl-devel  \
     centos-release-scl  \
-    rh-python38  \
-    rh-python38-python-devel \
     which
 
 # Install protoc
@@ -24,7 +22,7 @@ RUN unzip protoc-3.15.8-linux-x86_64.zip -d /usr/local/
 # Install Rust
 SHELL ["/bin/bash", "-c"]
 RUN curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- --no-modify-path --default-toolchain none -y
-ENV PATH /opt/rh/rh-python38/root/usr/bin:/usr/local/bin:/root/.cargo/bin/:$PATH
+ENV PATH /usr/local/bin:/root/.cargo/bin/:$PATH
 
 # Build the project in release mode.
 RUN --mount=target=.,rw \
@@ -43,8 +41,6 @@ RUN yum install -y epel-release \
     openssl \
     openssl-devel  \
     centos-release-scl  \
-    rh-python38  \
-    rh-python38-python-devel \
     which
 
 WORKDIR /greptime

docker/buildx/ubuntu/Dockerfile

@@ -7,10 +7,8 @@ ARG OUTPUT_DIR
 ENV LANG en_US.utf8
 WORKDIR /greptimedb
 
-# Add PPA for Python 3.10.
 RUN apt-get update && \
-    DEBIAN_FRONTEND=noninteractive apt-get install -y software-properties-common && \
-    add-apt-repository ppa:deadsnakes/ppa -y
+    DEBIAN_FRONTEND=noninteractive apt-get install -y software-properties-common
 
 # Install dependencies.
 RUN --mount=type=cache,target=/var/cache/apt \
@@ -20,10 +18,7 @@ RUN --mount=type=cache,target=/var/cache/apt \
     curl \
     git \
     build-essential \
-    pkg-config \
-    python3.10 \
-    python3.10-dev \
-    python3-pip
+    pkg-config
 
 # Install Rust.
 SHELL ["/bin/bash", "-c"]
@@ -46,15 +41,8 @@ ARG OUTPUT_DIR
 
 RUN apt-get update && DEBIAN_FRONTEND=noninteractive apt-get \
     -y install ca-certificates \
-    python3.10 \
-    python3.10-dev \
-    python3-pip \
     curl
 
-COPY ./docker/python/requirements.txt /etc/greptime/requirements.txt
-
-RUN python3 -m pip install -r /etc/greptime/requirements.txt
-
 WORKDIR /greptime
 COPY --from=builder /out/target/${OUTPUT_DIR}/greptime /greptime/bin/
 ENV PATH /greptime/bin/:$PATH

docker/ci/centos/Dockerfile

@@ -7,9 +7,7 @@ RUN sed -i s/^#.*baseurl=http/baseurl=http/g /etc/yum.repos.d/*.repo
 RUN yum install -y epel-release \
     openssl \
     openssl-devel  \
-    centos-release-scl  \
-    rh-python38  \
-    rh-python38-python-devel
+    centos-release-scl
 
 ARG TARGETARCH
 

docker/ci/ubuntu/Dockerfile

@@ -8,15 +8,8 @@ ARG TARGET_BIN=greptime
 
 RUN apt-get update && DEBIAN_FRONTEND=noninteractive apt-get install -y \
     ca-certificates \
-    python3.10 \
-    python3.10-dev \
-    python3-pip \
     curl
 
-COPY $DOCKER_BUILD_ROOT/docker/python/requirements.txt /etc/greptime/requirements.txt
-
-RUN python3 -m pip install -r /etc/greptime/requirements.txt
-
 ARG TARGETARCH
 
 ADD $TARGETARCH/$TARGET_BIN /greptime/bin/

docker/dev-builder/android/Dockerfile

@@ -9,16 +9,20 @@ RUN cp ${NDK_ROOT}/toolchains/llvm/prebuilt/linux-x86_64/lib64/clang/14.0.7/lib/
 # Install dependencies.
 RUN apt-get update && apt-get install -y \
     libssl-dev \
-    protobuf-compiler \
     curl \
     git \
+    unzip \
     build-essential \
-    pkg-config \
-    python3 \
-    python3-dev \
-    python3-pip \
-    && pip3 install --upgrade pip \
-    && pip3 install pyarrow
+    pkg-config
+
+# Install protoc
+ARG PROTOBUF_VERSION=29.3
+
+RUN curl -OL https://github.com/protocolbuffers/protobuf/releases/download/v${PROTOBUF_VERSION}/protoc-${PROTOBUF_VERSION}-linux-x86_64.zip && \
+    unzip protoc-${PROTOBUF_VERSION}-linux-x86_64.zip -d protoc3;
+    
+RUN mv protoc3/bin/* /usr/local/bin/
+RUN mv protoc3/include/* /usr/local/include/
 
 # Trust workdir
 RUN git config --global --add safe.directory /greptimedb

docker/dev-builder/centos/Dockerfile

@@ -12,18 +12,21 @@ RUN yum install -y epel-release  \
     openssl \
     openssl-devel  \
     centos-release-scl  \
-    rh-python38  \
-    rh-python38-python-devel \
     which
 
 # Install protoc
-RUN curl -LO https://github.com/protocolbuffers/protobuf/releases/download/v3.15.8/protoc-3.15.8-linux-x86_64.zip
-RUN unzip protoc-3.15.8-linux-x86_64.zip -d /usr/local/
+ARG PROTOBUF_VERSION=29.3
+
+RUN curl -OL https://github.com/protocolbuffers/protobuf/releases/download/v${PROTOBUF_VERSION}/protoc-${PROTOBUF_VERSION}-linux-x86_64.zip && \
+    unzip protoc-${PROTOBUF_VERSION}-linux-x86_64.zip -d protoc3;
+    
+RUN mv protoc3/bin/* /usr/local/bin/
+RUN mv protoc3/include/* /usr/local/include/
 
 # Install Rust
 SHELL ["/bin/bash", "-c"]
 RUN curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- --no-modify-path --default-toolchain none -y
-ENV PATH /opt/rh/rh-python38/root/usr/bin:/usr/local/bin:/root/.cargo/bin/:$PATH
+ENV PATH /usr/local/bin:/root/.cargo/bin/:$PATH
 
 # Install Rust toolchains.
 ARG RUST_TOOLCHAIN

docker/dev-builder/ubuntu/Dockerfile

@@ -6,11 +6,8 @@ ARG DOCKER_BUILD_ROOT=.
 ENV LANG en_US.utf8
 WORKDIR /greptimedb
 
-# Add PPA for Python 3.10.
 RUN apt-get update && \
-    DEBIAN_FRONTEND=noninteractive apt-get install -y software-properties-common && \
-    add-apt-repository ppa:deadsnakes/ppa -y
-
+    DEBIAN_FRONTEND=noninteractive apt-get install -y software-properties-common
 # Install dependencies.
 RUN apt-get update && DEBIAN_FRONTEND=noninteractive apt-get install -y \
     libssl-dev \
@@ -20,39 +17,24 @@ RUN apt-get update && DEBIAN_FRONTEND=noninteractive apt-get install -y \
     ca-certificates \
     git \
     build-essential \
-    pkg-config \
-    python3.10 \
-    python3.10-dev
+    pkg-config
 
 ARG TARGETPLATFORM
 RUN echo "target platform: $TARGETPLATFORM"
 
+ARG PROTOBUF_VERSION=29.3
+
 # Install protobuf, because the one in the apt is too old (v3.12).
 RUN if [ "$TARGETPLATFORM" = "linux/arm64" ]; then \
-    curl -OL https://github.com/protocolbuffers/protobuf/releases/download/v29.1/protoc-29.1-linux-aarch_64.zip && \
-    unzip protoc-29.1-linux-aarch_64.zip -d protoc3; \
+    curl -OL https://github.com/protocolbuffers/protobuf/releases/download/v${PROTOBUF_VERSION}/protoc-${PROTOBUF_VERSION}-linux-aarch_64.zip && \
+    unzip protoc-${PROTOBUF_VERSION}-linux-aarch_64.zip -d protoc3; \
 elif [ "$TARGETPLATFORM" = "linux/amd64" ]; then \
-    curl -OL https://github.com/protocolbuffers/protobuf/releases/download/v29.1/protoc-29.1-linux-x86_64.zip && \
-    unzip protoc-29.1-linux-x86_64.zip -d protoc3; \
+    curl -OL https://github.com/protocolbuffers/protobuf/releases/download/v${PROTOBUF_VERSION}/protoc-${PROTOBUF_VERSION}-linux-x86_64.zip && \
+    unzip protoc-${PROTOBUF_VERSION}-linux-x86_64.zip -d protoc3; \
 fi
 RUN mv protoc3/bin/* /usr/local/bin/
 RUN mv protoc3/include/* /usr/local/include/
 
-# https://github.com/GreptimeTeam/greptimedb/actions/runs/10935485852/job/30357457188#step:3:7106
-# `aws-lc-sys` require gcc >= 10.3.0 to work, hence alias to use gcc-10
-RUN apt-get remove -y gcc-9 g++-9 cpp-9 && \
-    apt-get install -y gcc-10 g++-10 cpp-10 make cmake && \
-    ln -sf /usr/bin/gcc-10 /usr/bin/gcc && ln -sf /usr/bin/g++-10 /usr/bin/g++ && \
-    ln -sf /usr/bin/gcc-10 /usr/bin/cc && \
-    ln -sf /usr/bin/g++-10 /usr/bin/cpp && ln -sf /usr/bin/g++-10 /usr/bin/c++ && \
-    cc --version && gcc --version && g++ --version && cpp --version && c++ --version
-
-# Remove Python 3.8 and install pip.
-RUN apt-get -y purge python3.8 && \
-    apt-get -y autoremove && \
-    ln -s /usr/bin/python3.10 /usr/bin/python3 && \
-    curl -sS https://bootstrap.pypa.io/get-pip.py | python3.10
-
 # Silence all `safe.directory` warnings, to avoid the "detect dubious repository" error when building with submodules.
 # Disabling the safe directory check here won't pose extra security issues, because in our usage for this dev build
 # image, we use it solely on our own environment (that github action's VM, or ECS created dynamically by ourselves),
@@ -65,10 +47,6 @@ RUN apt-get -y purge python3.8 && \
 # it can be a different user that have prepared the submodules.
 RUN git config --global --add safe.directory '*'
 
-# Install Python dependencies.
-COPY $DOCKER_BUILD_ROOT/docker/python/requirements.txt /etc/greptime/requirements.txt
-RUN python3 -m pip install -r /etc/greptime/requirements.txt
-
 # Install Rust.
 SHELL ["/bin/bash", "-c"]
 RUN curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- --no-modify-path --default-toolchain none -y

docker/dev-builder/ubuntu/Dockerfile-18.10

@@ -21,7 +21,7 @@ RUN apt-get update && DEBIAN_FRONTEND=noninteractive apt-get install -y \
     pkg-config
 
 # Install protoc.
-ENV PROTOC_VERSION=25.1
+ENV PROTOC_VERSION=29.3
 RUN if [ "$(uname -m)" = "x86_64" ]; then \
         PROTOC_ZIP=protoc-${PROTOC_VERSION}-linux-x86_64.zip; \
     elif [ "$(uname -m)" = "aarch64" ]; then \

docker/docker-compose/cluster-with-etcd.yaml

@@ -39,14 +39,16 @@ services:
     container_name: metasrv
     ports:
       - 3002:3002
+      - 3000:3000
     command:
       - metasrv
       - start
       - --bind-addr=0.0.0.0:3002
       - --server-addr=metasrv:3002
       - --store-addrs=etcd0:2379
+      - --http-addr=0.0.0.0:3000
     healthcheck:
-      test: [ "CMD", "curl", "-f", "http://metasrv:3002/health" ]
+      test: [ "CMD", "curl", "-f", "http://metasrv:3000/health" ]
       interval: 5s
       timeout: 3s
       retries: 5
@@ -73,10 +75,10 @@ services:
     volumes:
       - /tmp/greptimedb-cluster-docker-compose/datanode0:/tmp/greptimedb
     healthcheck:
-      test: [ "CMD", "curl", "-f", "http://datanode0:5000/health" ]
+      test: [ "CMD", "curl", "-fv", "http://datanode0:5000/health" ]
       interval: 5s
       timeout: 3s
-      retries: 5
+      retries: 10
     depends_on:
       metasrv:
         condition: service_healthy
@@ -115,6 +117,7 @@ services:
     container_name: flownode0
     ports:
       - 4004:4004
+      - 4005:4005
     command:
       - flownode
       - start
@@ -122,9 +125,15 @@ services:
       - --metasrv-addrs=metasrv:3002
       - --rpc-addr=0.0.0.0:4004
       - --rpc-hostname=flownode0:4004
+      - --http-addr=0.0.0.0:4005
     depends_on:
       frontend0:
         condition: service_healthy
+    healthcheck:
+      test: [ "CMD", "curl", "-f", "http://flownode0:4005/health" ]
+      interval: 5s
+      timeout: 3s
+      retries: 5
     networks:
       - greptimedb
 

docker/python/requirements.txt

@@ -1,5 +0,0 @@
-numpy>=1.24.2
-pandas>=1.5.3
-pyarrow>=11.0.0
-requests>=2.28.2
-scipy>=1.10.1

flake.lock

@@ -0,0 +1,100 @@
+{
+  "nodes": {
+    "fenix": {
+      "inputs": {
+        "nixpkgs": [
+          "nixpkgs"
+        ],
+        "rust-analyzer-src": "rust-analyzer-src"
+      },
+      "locked": {
+        "lastModified": 1737613896,
+        "narHash": "sha256-ldqXIglq74C7yKMFUzrS9xMT/EVs26vZpOD68Sh7OcU=",
+        "owner": "nix-community",
+        "repo": "fenix",
+        "rev": "303a062fdd8e89f233db05868468975d17855d80",
+        "type": "github"
+      },
+      "original": {
+        "owner": "nix-community",
+        "repo": "fenix",
+        "type": "github"
+      }
+    },
+    "flake-utils": {
+      "inputs": {
+        "systems": "systems"
+      },
+      "locked": {
+        "lastModified": 1731533236,
+        "narHash": "sha256-l0KFg5HjrsfsO/JpG+r7fRrqm12kzFHyUHqHCVpMMbI=",
+        "owner": "numtide",
+        "repo": "flake-utils",
+        "rev": "11707dc2f618dd54ca8739b309ec4fc024de578b",
+        "type": "github"
+      },
+      "original": {
+        "owner": "numtide",
+        "repo": "flake-utils",
+        "type": "github"
+      }
+    },
+    "nixpkgs": {
+      "locked": {
+        "lastModified": 1737569578,
+        "narHash": "sha256-6qY0pk2QmUtBT9Mywdvif0i/CLVgpCjMUn6g9vB+f3M=",
+        "owner": "NixOS",
+        "repo": "nixpkgs",
+        "rev": "47addd76727f42d351590c905d9d1905ca895b82",
+        "type": "github"
+      },
+      "original": {
+        "owner": "NixOS",
+        "ref": "nixos-24.11",
+        "repo": "nixpkgs",
+        "type": "github"
+      }
+    },
+    "root": {
+      "inputs": {
+        "fenix": "fenix",
+        "flake-utils": "flake-utils",
+        "nixpkgs": "nixpkgs"
+      }
+    },
+    "rust-analyzer-src": {
+      "flake": false,
+      "locked": {
+        "lastModified": 1737581772,
+        "narHash": "sha256-t1P2Pe3FAX9TlJsCZbmJ3wn+C4qr6aSMypAOu8WNsN0=",
+        "owner": "rust-lang",
+        "repo": "rust-analyzer",
+        "rev": "582af7ee9c8d84f5d534272fc7de9f292bd849be",
+        "type": "github"
+      },
+      "original": {
+        "owner": "rust-lang",
+        "ref": "nightly",
+        "repo": "rust-analyzer",
+        "type": "github"
+      }
+    },
+    "systems": {
+      "locked": {
+        "lastModified": 1681028828,
+        "narHash": "sha256-Vy1rq5AaRuLzOxct8nz4T6wlgyUR7zLU309k9mBC768=",
+        "owner": "nix-systems",
+        "repo": "default",
+        "rev": "da67096a3b9bf56a91d16901293e51ba5b49a27e",
+        "type": "github"
+      },
+      "original": {
+        "owner": "nix-systems",
+        "repo": "default",
+        "type": "github"
+      }
+    }
+  },
+  "root": "root",
+  "version": 7
+}

flake.nix

@@ -0,0 +1,56 @@
+{
+  description = "Development environment flake";
+
+  inputs = {
+    nixpkgs.url = "github:NixOS/nixpkgs/nixos-24.11";
+    fenix = {
+      url = "github:nix-community/fenix";
+      inputs.nixpkgs.follows = "nixpkgs";
+    };
+    flake-utils.url = "github:numtide/flake-utils";
+  };
+
+  outputs = { self, nixpkgs, fenix, flake-utils }:
+    flake-utils.lib.eachDefaultSystem (system:
+      let
+        pkgs = nixpkgs.legacyPackages.${system};
+        buildInputs = with pkgs; [
+          libgit2
+          libz
+        ];
+        lib = nixpkgs.lib;
+        rustToolchain = fenix.packages.${system}.fromToolchainName {
+          name = (lib.importTOML ./rust-toolchain.toml).toolchain.channel;
+          sha256 = "sha256-f/CVA1EC61EWbh0SjaRNhLL0Ypx2ObupbzigZp8NmL4=";
+        };
+      in
+      {
+        devShells.default = pkgs.mkShell {
+          nativeBuildInputs = with pkgs; [
+            pkg-config
+            git
+            clang
+            gcc
+            protobuf
+            gnumake
+            mold
+            (rustToolchain.withComponents [
+              "cargo"
+              "clippy"
+              "rust-src"
+              "rustc"
+              "rustfmt"
+              "rust-analyzer"
+              "llvm-tools"
+            ])
+            cargo-nextest
+            cargo-llvm-cov
+            taplo
+            curl
+            gnuplot ## for cargo bench
+          ];
+
+          LD_LIBRARY_PATH = pkgs.lib.makeLibraryPath buildInputs;
+        };
+      });
+}

rust-toolchain.toml

@@ -1,3 +1,2 @@
 [toolchain]
-channel = "nightly-2024-10-19"
-components = ["rust-analyzer"]
+channel = "nightly-2024-12-25"

scripts/check-snafu.py

@@ -14,6 +14,7 @@
 
 import os
 import re
+from multiprocessing import Pool
 
 
 def find_rust_files(directory):
@@ -33,13 +34,11 @@ def extract_branch_names(file_content):
     return pattern.findall(file_content)
 
 
-def check_snafu_in_files(branch_name, rust_files):
+def check_snafu_in_files(branch_name, rust_files_content):
     branch_name_snafu = f"{branch_name}Snafu"
-    for rust_file in rust_files:
-        with open(rust_file, "r") as file:
-            content = file.read()
-            if branch_name_snafu in content:
-                return True
+    for content in rust_files_content.values():
+        if branch_name_snafu in content:
+            return True
     return False
 
 
@@ -49,21 +48,24 @@ def main():
 
     for error_file in error_files:
         with open(error_file, "r") as file:
-            content = file.read()
-            branch_names.extend(extract_branch_names(content))
+            branch_names.extend(extract_branch_names(file.read()))
 
-    unused_snafu = [
-        branch_name
-        for branch_name in branch_names
-        if not check_snafu_in_files(branch_name, other_rust_files)
-    ]
+    # Read all rust files into memory once
+    rust_files_content = {}
+    for rust_file in other_rust_files:
+        with open(rust_file, "r") as file:
+            rust_files_content[rust_file] = file.read()
+
+    with Pool() as pool:
+        results = pool.starmap(
+            check_snafu_in_files, [(bn, rust_files_content) for bn in branch_names]
+        )
+    unused_snafu = [bn for bn, found in zip(branch_names, results) if not found]
 
     if unused_snafu:
         print("Unused error variants:")
         for name in unused_snafu:
             print(name)
-
-    if unused_snafu:
         raise SystemExit(1)
 
 

shell.nix

@@ -1,27 +0,0 @@
-let
-  nixpkgs = fetchTarball "https://github.com/NixOS/nixpkgs/tarball/nixos-unstable";
-  fenix = import (fetchTarball "https://github.com/nix-community/fenix/archive/main.tar.gz") {};
-  pkgs = import nixpkgs { config = {}; overlays = []; };
-in
-
-pkgs.mkShell rec {
-  nativeBuildInputs = with pkgs; [
-    pkg-config
-    git
-    clang
-    gcc
-    protobuf
-    mold
-    (fenix.fromToolchainFile {
-      dir = ./.;
-    })
-    cargo-nextest
-    taplo
-  ];
-
-  buildInputs = with pkgs; [
-    libgit2
-  ];
-
-  LD_LIBRARY_PATH = pkgs.lib.makeLibraryPath buildInputs;
-}

src/api/src/error.rs

@@ -33,7 +33,7 @@ pub enum Error {
         #[snafu(implicit)]
         location: Location,
         #[snafu(source)]
-        error: prost::DecodeError,
+        error: prost::UnknownEnumValue,
     },
 
     #[snafu(display("Failed to create column datatype from {:?}", from))]

src/api/src/helper.rs

@@ -86,7 +86,7 @@ impl ColumnDataTypeWrapper {
 
     /// Get a tuple of ColumnDataType and ColumnDataTypeExtension.
     pub fn to_parts(&self) -> (ColumnDataType, Option<ColumnDataTypeExtension>) {
-        (self.datatype, self.datatype_ext.clone())
+        (self.datatype, self.datatype_ext)
     }
 }
 
@@ -685,14 +685,18 @@ pub fn pb_values_to_vector_ref(data_type: &ConcreteDataType, values: Values) ->
             IntervalType::YearMonth(_) => Arc::new(IntervalYearMonthVector::from_vec(
                 values.interval_year_month_values,
             )),
-            IntervalType::DayTime(_) => Arc::new(IntervalDayTimeVector::from_vec(
-                values.interval_day_time_values,
+            IntervalType::DayTime(_) => Arc::new(IntervalDayTimeVector::from_iter_values(
+                values
+                    .interval_day_time_values
+                    .iter()
+                    .map(|x| IntervalDayTime::from_i64(*x).into()),
             )),
             IntervalType::MonthDayNano(_) => {
                 Arc::new(IntervalMonthDayNanoVector::from_iter_values(
-                    values.interval_month_day_nano_values.iter().map(|x| {
-                        IntervalMonthDayNano::new(x.months, x.days, x.nanoseconds).to_i128()
-                    }),
+                    values
+                        .interval_month_day_nano_values
+                        .iter()
+                        .map(|x| IntervalMonthDayNano::new(x.months, x.days, x.nanoseconds).into()),
                 ))
             }
         },
@@ -1495,14 +1499,22 @@ mod tests {
             column.values.as_ref().unwrap().interval_year_month_values
         );
 
-        let vector = Arc::new(IntervalDayTimeVector::from_vec(vec![4, 5, 6]));
+        let vector = Arc::new(IntervalDayTimeVector::from_vec(vec![
+            IntervalDayTime::new(0, 4).into(),
+            IntervalDayTime::new(0, 5).into(),
+            IntervalDayTime::new(0, 6).into(),
+        ]));
         push_vals(&mut column, 3, vector);
         assert_eq!(
             vec![4, 5, 6],
             column.values.as_ref().unwrap().interval_day_time_values
         );
 
-        let vector = Arc::new(IntervalMonthDayNanoVector::from_vec(vec![7, 8, 9]));
+        let vector = Arc::new(IntervalMonthDayNanoVector::from_vec(vec![
+            IntervalMonthDayNano::new(0, 0, 7).into(),
+            IntervalMonthDayNano::new(0, 0, 8).into(),
+            IntervalMonthDayNano::new(0, 0, 9).into(),
+        ]));
         let len = vector.len();
         push_vals(&mut column, 3, vector);
         (0..len).for_each(|i| {

src/api/src/v1/column_def.rs

@@ -34,10 +34,8 @@ const SKIPPING_INDEX_GRPC_KEY: &str = "skipping_index";
 
 /// Tries to construct a `ColumnSchema` from the given  `ColumnDef`.
 pub fn try_as_column_schema(column_def: &ColumnDef) -> Result<ColumnSchema> {
-    let data_type = ColumnDataTypeWrapper::try_new(
-        column_def.data_type,
-        column_def.datatype_extension.clone(),
-    )?;
+    let data_type =
+        ColumnDataTypeWrapper::try_new(column_def.data_type, column_def.datatype_extension)?;
 
     let constraint = if column_def.default_constraint.is_empty() {
         None
@@ -57,13 +55,13 @@ pub fn try_as_column_schema(column_def: &ColumnDef) -> Result<ColumnSchema> {
     }
     if let Some(options) = column_def.options.as_ref() {
         if let Some(fulltext) = options.options.get(FULLTEXT_GRPC_KEY) {
-            metadata.insert(FULLTEXT_KEY.to_string(), fulltext.clone());
+            metadata.insert(FULLTEXT_KEY.to_string(), fulltext.to_owned());
         }
         if let Some(inverted_index) = options.options.get(INVERTED_INDEX_GRPC_KEY) {
-            metadata.insert(INVERTED_INDEX_KEY.to_string(), inverted_index.clone());
+            metadata.insert(INVERTED_INDEX_KEY.to_string(), inverted_index.to_owned());
         }
         if let Some(skipping_index) = options.options.get(SKIPPING_INDEX_GRPC_KEY) {
-            metadata.insert(SKIPPING_INDEX_KEY.to_string(), skipping_index.clone());
+            metadata.insert(SKIPPING_INDEX_KEY.to_string(), skipping_index.to_owned());
         }
     }
 
@@ -82,7 +80,7 @@ pub fn options_from_column_schema(column_schema: &ColumnSchema) -> Option<Column
     if let Some(fulltext) = column_schema.metadata().get(FULLTEXT_KEY) {
         options
             .options
-            .insert(FULLTEXT_GRPC_KEY.to_string(), fulltext.clone());
+            .insert(FULLTEXT_GRPC_KEY.to_string(), fulltext.to_owned());
     }
     if let Some(inverted_index) = column_schema.metadata().get(INVERTED_INDEX_KEY) {
         options
@@ -102,7 +100,7 @@ pub fn options_from_column_schema(column_schema: &ColumnSchema) -> Option<Column
 pub fn contains_fulltext(options: &Option<ColumnOptions>) -> bool {
     options
         .as_ref()
-        .map_or(false, |o| o.options.contains_key(FULLTEXT_GRPC_KEY))
+        .is_some_and(|o| o.options.contains_key(FULLTEXT_GRPC_KEY))
 }
 
 /// Tries to construct a `ColumnOptions` from the given `FulltextOptions`.
@@ -181,14 +179,14 @@ mod tests {
         let options = options_from_column_schema(&schema);
         assert!(options.is_none());
 
-        let schema = ColumnSchema::new("test", ConcreteDataType::string_datatype(), true)
+        let mut schema = ColumnSchema::new("test", ConcreteDataType::string_datatype(), true)
             .with_fulltext_options(FulltextOptions {
                 enable: true,
                 analyzer: FulltextAnalyzer::English,
                 case_sensitive: false,
             })
-            .unwrap()
-            .set_inverted_index(true);
+            .unwrap();
+        schema.set_inverted_index(true);
         let options = options_from_column_schema(&schema).unwrap();
         assert_eq!(
             options.options.get(FULLTEXT_GRPC_KEY).unwrap(),

src/catalog/src/error.rs

@@ -122,13 +122,6 @@ pub enum Error {
         source: BoxedError,
     },
 
-    #[snafu(display("Failed to re-compile script due to internal error"))]
-    CompileScriptInternal {
-        #[snafu(implicit)]
-        location: Location,
-        source: BoxedError,
-    },
-
     #[snafu(display("Failed to create table, table info: {}", table_info))]
     CreateTable {
         table_info: String,
@@ -343,9 +336,7 @@ impl ErrorExt for Error {
             Error::DecodePlan { source, .. } => source.status_code(),
             Error::InvalidTableInfoInCatalog { source, .. } => source.status_code(),
 
-            Error::CompileScriptInternal { source, .. } | Error::Internal { source, .. } => {
-                source.status_code()
-            }
+            Error::Internal { source, .. } => source.status_code(),
 
             Error::QueryAccessDenied { .. } => StatusCode::AccessDenied,
             Error::Datafusion { error, .. } => datafusion_status_code::<Self>(error, None),

src/catalog/src/kvbackend/client.rs

@@ -303,7 +303,7 @@ impl KvBackend for CachedKvBackend {
             .lock()
             .unwrap()
             .as_ref()
-            .map_or(false, |v| !self.validate_version(*v))
+            .is_some_and(|v| !self.validate_version(*v))
         {
             self.cache.invalidate(key).await;
         }

src/catalog/src/lib.rs

@@ -41,6 +41,7 @@ pub mod information_schema {
 }
 
 pub mod table_source;
+
 #[async_trait::async_trait]
 pub trait CatalogManager: Send + Sync {
     fn as_any(&self) -> &dyn Any;

src/catalog/src/system_schema/information_schema/cluster_info.rs

@@ -64,6 +64,7 @@ const INIT_CAPACITY: usize = 42;
 /// - `uptime`: the uptime of the peer.
 /// - `active_time`: the time since the last activity of the peer.
 ///
+#[derive(Debug)]
 pub(super) struct InformationSchemaClusterInfo {
     schema: SchemaRef,
     catalog_manager: Weak<dyn CatalogManager>,

src/catalog/src/system_schema/information_schema/columns.rs

@@ -45,6 +45,7 @@ use crate::error::{
 use crate::information_schema::Predicates;
 use crate::CatalogManager;
 
+#[derive(Debug)]
 pub(super) struct InformationSchemaColumns {
     schema: SchemaRef,
     catalog_name: String,

src/catalog/src/system_schema/information_schema/flows.rs

@@ -61,7 +61,7 @@ pub const FLOWNODE_IDS: &str = "flownode_ids";
 pub const OPTIONS: &str = "options";
 
 /// The `information_schema.flows` to provides information about flows in databases.
-///
+#[derive(Debug)]
 pub(super) struct InformationSchemaFlows {
     schema: SchemaRef,
     catalog_name: String,

src/catalog/src/system_schema/information_schema/key_column_usage.rs

@@ -58,8 +58,11 @@ pub(crate) const TIME_INDEX_CONSTRAINT_NAME: &str = "TIME INDEX";
 pub(crate) const INVERTED_INDEX_CONSTRAINT_NAME: &str = "INVERTED INDEX";
 /// Fulltext index constraint name
 pub(crate) const FULLTEXT_INDEX_CONSTRAINT_NAME: &str = "FULLTEXT INDEX";
+/// Skipping index constraint name
+pub(crate) const SKIPPING_INDEX_CONSTRAINT_NAME: &str = "SKIPPING INDEX";
 
 /// The virtual table implementation for `information_schema.KEY_COLUMN_USAGE`.
+#[derive(Debug)]
 pub(super) struct InformationSchemaKeyColumnUsage {
     schema: SchemaRef,
     catalog_name: String,
@@ -225,6 +228,12 @@ impl InformationSchemaKeyColumnUsageBuilder {
                 let keys = &table_info.meta.primary_key_indices;
                 let schema = table.schema();
 
+                // For compatibility, use primary key columns as inverted index columns.
+                let pk_as_inverted_index = !schema
+                    .column_schemas()
+                    .iter()
+                    .any(|c| c.has_inverted_index_key());
+
                 for (idx, column) in schema.column_schemas().iter().enumerate() {
                     let mut constraints = vec![];
                     if column.is_time_index() {
@@ -242,14 +251,20 @@ impl InformationSchemaKeyColumnUsageBuilder {
                     // TODO(dimbtp): foreign key constraint not supported yet
                     if keys.contains(&idx) {
                         constraints.push(PRI_CONSTRAINT_NAME);
+
+                        if pk_as_inverted_index {
+                            constraints.push(INVERTED_INDEX_CONSTRAINT_NAME);
+                        }
                     }
                     if column.is_inverted_indexed() {
                         constraints.push(INVERTED_INDEX_CONSTRAINT_NAME);
                     }
-
-                    if column.has_fulltext_index_key() {
+                    if column.is_fulltext_indexed() {
                         constraints.push(FULLTEXT_INDEX_CONSTRAINT_NAME);
                     }
+                    if column.is_skipping_indexed() {
+                        constraints.push(SKIPPING_INDEX_CONSTRAINT_NAME);
+                    }
 
                     if !constraints.is_empty() {
                         let aggregated_constraints = constraints.join(", ");

src/catalog/src/system_schema/information_schema/partitions.rs

@@ -59,6 +59,7 @@ const INIT_CAPACITY: usize = 42;
 /// The `PARTITIONS` table provides information about partitioned tables.
 /// See https://dev.mysql.com/doc/refman/8.0/en/information-schema-partitions-table.html
 /// We provide an extral column `greptime_partition_id` for GreptimeDB region id.
+#[derive(Debug)]
 pub(super) struct InformationSchemaPartitions {
     schema: SchemaRef,
     catalog_name: String,

src/catalog/src/system_schema/information_schema/procedure_info.rs

@@ -56,7 +56,7 @@ const INIT_CAPACITY: usize = 42;
 /// - `end_time`: the ending execution time of the procedure.
 /// - `status`: the status of the procedure.
 /// - `lock_keys`: the lock keys of the procedure.
-///
+#[derive(Debug)]
 pub(super) struct InformationSchemaProcedureInfo {
     schema: SchemaRef,
     catalog_manager: Weak<dyn CatalogManager>,

src/catalog/src/system_schema/information_schema/region_peers.rs

@@ -59,7 +59,7 @@ const INIT_CAPACITY: usize = 42;
 /// - `is_leader`: whether the peer is the leader
 /// - `status`: the region status, `ALIVE` or `DOWNGRADED`.
 /// - `down_seconds`: the duration of being offline, in seconds.
-///
+#[derive(Debug)]
 pub(super) struct InformationSchemaRegionPeers {
     schema: SchemaRef,
     catalog_name: String,

src/catalog/src/system_schema/information_schema/region_statistics.rs

@@ -63,7 +63,7 @@ const INIT_CAPACITY: usize = 42;
 /// - `index_size`: The sst index files size in bytes.
 /// - `engine`: The engine type.
 /// - `region_role`: The region role.
-///
+#[derive(Debug)]
 pub(super) struct InformationSchemaRegionStatistics {
     schema: SchemaRef,
     catalog_manager: Weak<dyn CatalogManager>,

src/catalog/src/system_schema/information_schema/runtime_metrics.rs

@@ -38,6 +38,7 @@ use store_api::storage::{ScanRequest, TableId};
 use super::{InformationTable, RUNTIME_METRICS};
 use crate::error::{CreateRecordBatchSnafu, InternalSnafu, Result};
 
+#[derive(Debug)]
 pub(super) struct InformationSchemaMetrics {
     schema: SchemaRef,
 }

src/catalog/src/system_schema/information_schema/schemata.rs

@@ -49,6 +49,7 @@ pub const SCHEMA_OPTS: &str = "options";
 const INIT_CAPACITY: usize = 42;
 
 /// The `information_schema.schemata` table implementation.
+#[derive(Debug)]
 pub(super) struct InformationSchemaSchemata {
     schema: SchemaRef,
     catalog_name: String,

src/catalog/src/system_schema/information_schema/table_constraints.rs

@@ -43,6 +43,7 @@ use crate::information_schema::Predicates;
 use crate::CatalogManager;
 
 /// The `TABLE_CONSTRAINTS` table describes which tables have constraints.
+#[derive(Debug)]
 pub(super) struct InformationSchemaTableConstraints {
     schema: SchemaRef,
     catalog_name: String,

src/catalog/src/system_schema/information_schema/tables.rs

@@ -71,6 +71,7 @@ const TABLE_ID: &str = "table_id";
 pub const ENGINE: &str = "engine";
 const INIT_CAPACITY: usize = 42;
 
+#[derive(Debug)]
 pub(super) struct InformationSchemaTables {
     schema: SchemaRef,
     catalog_name: String,

src/catalog/src/system_schema/information_schema/views.rs

@@ -54,6 +54,7 @@ pub const CHARACTER_SET_CLIENT: &str = "character_set_client";
 pub const COLLATION_CONNECTION: &str = "collation_connection";
 
 /// The `information_schema.views` to provides information about views in databases.
+#[derive(Debug)]
 pub(super) struct InformationSchemaViews {
     schema: SchemaRef,
     catalog_name: String,

src/catalog/src/system_schema/memory_table.rs

@@ -33,6 +33,7 @@ use super::SystemTable;
 use crate::error::{CreateRecordBatchSnafu, InternalSnafu, Result};
 
 /// A memory table with specified schema and columns.
+#[derive(Debug)]
 pub(crate) struct MemoryTable {
     pub(crate) table_id: TableId,
     pub(crate) table_name: &'static str,

src/catalog/src/system_schema/pg_catalog.rs

@@ -14,6 +14,7 @@
 
 mod pg_catalog_memory_table;
 mod pg_class;
+mod pg_database;
 mod pg_namespace;
 mod table_names;
 
@@ -26,6 +27,7 @@ use lazy_static::lazy_static;
 use paste::paste;
 use pg_catalog_memory_table::get_schema_columns;
 use pg_class::PGClass;
+use pg_database::PGDatabase;
 use pg_namespace::PGNamespace;
 use session::context::{Channel, QueryContext};
 use table::TableRef;
@@ -113,6 +115,10 @@ impl PGCatalogProvider {
             PG_CLASS.to_string(),
             self.build_table(PG_CLASS).expect(PG_NAMESPACE),
         );
+        tables.insert(
+            PG_DATABASE.to_string(),
+            self.build_table(PG_DATABASE).expect(PG_DATABASE),
+        );
         self.tables = tables;
     }
 }
@@ -135,6 +141,11 @@ impl SystemSchemaProviderInner for PGCatalogProvider {
                 self.catalog_manager.clone(),
                 self.namespace_oid_map.clone(),
             ))),
+            table_names::PG_DATABASE => Some(Arc::new(PGDatabase::new(
+                self.catalog_name.clone(),
+                self.catalog_manager.clone(),
+                self.namespace_oid_map.clone(),
+            ))),
             _ => None,
         }
     }

src/catalog/src/system_schema/pg_catalog/pg_class.rs

@@ -12,6 +12,7 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
+use std::fmt;
 use std::sync::{Arc, Weak};
 
 use arrow_schema::SchemaRef as ArrowSchemaRef;
@@ -100,6 +101,15 @@ impl PGClass {
     }
 }
 
+impl fmt::Debug for PGClass {
+    fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
+        f.debug_struct("PGClass")
+            .field("schema", &self.schema)
+            .field("catalog_name", &self.catalog_name)
+            .finish()
+    }
+}
+
 impl SystemTable for PGClass {
     fn table_id(&self) -> table::metadata::TableId {
         PG_CATALOG_PG_CLASS_TABLE_ID

src/catalog/src/system_schema/pg_catalog/pg_database.rs

@@ -0,0 +1,223 @@
+// Copyright 2023 Greptime Team
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//     http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
+use std::sync::{Arc, Weak};
+
+use arrow_schema::SchemaRef as ArrowSchemaRef;
+use common_catalog::consts::PG_CATALOG_PG_DATABASE_TABLE_ID;
+use common_error::ext::BoxedError;
+use common_recordbatch::adapter::RecordBatchStreamAdapter;
+use common_recordbatch::{DfSendableRecordBatchStream, RecordBatch};
+use datafusion::execution::TaskContext;
+use datafusion::physical_plan::stream::RecordBatchStreamAdapter as DfRecordBatchStreamAdapter;
+use datafusion::physical_plan::streaming::PartitionStream as DfPartitionStream;
+use datatypes::scalars::ScalarVectorBuilder;
+use datatypes::schema::{Schema, SchemaRef};
+use datatypes::value::Value;
+use datatypes::vectors::{StringVectorBuilder, UInt32VectorBuilder, VectorRef};
+use snafu::{OptionExt, ResultExt};
+use store_api::storage::ScanRequest;
+
+use super::pg_namespace::oid_map::PGNamespaceOidMapRef;
+use super::{query_ctx, OID_COLUMN_NAME, PG_DATABASE};
+use crate::error::{
+    CreateRecordBatchSnafu, InternalSnafu, Result, UpgradeWeakCatalogManagerRefSnafu,
+};
+use crate::information_schema::Predicates;
+use crate::system_schema::utils::tables::{string_column, u32_column};
+use crate::system_schema::SystemTable;
+use crate::CatalogManager;
+
+// === column name ===
+pub const DATNAME: &str = "datname";
+
+/// The initial capacity of the vector builders.
+const INIT_CAPACITY: usize = 42;
+
+/// The `pg_catalog.database` table implementation.
+pub(super) struct PGDatabase {
+    schema: SchemaRef,
+    catalog_name: String,
+    catalog_manager: Weak<dyn CatalogManager>,
+
+    // Workaround to convert schema_name to a numeric id
+    namespace_oid_map: PGNamespaceOidMapRef,
+}
+
+impl std::fmt::Debug for PGDatabase {
+    fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
+        f.debug_struct("PGDatabase")
+            .field("schema", &self.schema)
+            .field("catalog_name", &self.catalog_name)
+            .finish()
+    }
+}
+
+impl PGDatabase {
+    pub(super) fn new(
+        catalog_name: String,
+        catalog_manager: Weak<dyn CatalogManager>,
+        namespace_oid_map: PGNamespaceOidMapRef,
+    ) -> Self {
+        Self {
+            schema: Self::schema(),
+            catalog_name,
+            catalog_manager,
+            namespace_oid_map,
+        }
+    }
+
+    fn schema() -> SchemaRef {
+        Arc::new(Schema::new(vec![
+            u32_column(OID_COLUMN_NAME),
+            string_column(DATNAME),
+        ]))
+    }
+
+    fn builder(&self) -> PGCDatabaseBuilder {
+        PGCDatabaseBuilder::new(
+            self.schema.clone(),
+            self.catalog_name.clone(),
+            self.catalog_manager.clone(),
+            self.namespace_oid_map.clone(),
+        )
+    }
+}
+
+impl DfPartitionStream for PGDatabase {
+    fn schema(&self) -> &ArrowSchemaRef {
+        self.schema.arrow_schema()
+    }
+
+    fn execute(&self, _: Arc<TaskContext>) -> DfSendableRecordBatchStream {
+        let schema = self.schema.arrow_schema().clone();
+        let mut builder = self.builder();
+        Box::pin(DfRecordBatchStreamAdapter::new(
+            schema,
+            futures::stream::once(async move {
+                builder
+                    .make_database(None)
+                    .await
+                    .map(|x| x.into_df_record_batch())
+                    .map_err(Into::into)
+            }),
+        ))
+    }
+}
+
+impl SystemTable for PGDatabase {
+    fn table_id(&self) -> table::metadata::TableId {
+        PG_CATALOG_PG_DATABASE_TABLE_ID
+    }
+
+    fn table_name(&self) -> &'static str {
+        PG_DATABASE
+    }
+
+    fn schema(&self) -> SchemaRef {
+        self.schema.clone()
+    }
+
+    fn to_stream(
+        &self,
+        request: ScanRequest,
+    ) -> Result<common_recordbatch::SendableRecordBatchStream> {
+        let schema = self.schema.arrow_schema().clone();
+        let mut builder = self.builder();
+        let stream = Box::pin(DfRecordBatchStreamAdapter::new(
+            schema,
+            futures::stream::once(async move {
+                builder
+                    .make_database(Some(request))
+                    .await
+                    .map(|x| x.into_df_record_batch())
+                    .map_err(Into::into)
+            }),
+        ));
+        Ok(Box::pin(
+            RecordBatchStreamAdapter::try_new(stream)
+                .map_err(BoxedError::new)
+                .context(InternalSnafu)?,
+        ))
+    }
+}
+
+/// Builds the `pg_catalog.pg_database` table row by row
+/// `oid` use schema name as a workaround since we don't have numeric schema id.
+/// `nspname` is the schema name.
+struct PGCDatabaseBuilder {
+    schema: SchemaRef,
+    catalog_name: String,
+    catalog_manager: Weak<dyn CatalogManager>,
+    namespace_oid_map: PGNamespaceOidMapRef,
+
+    oid: UInt32VectorBuilder,
+    datname: StringVectorBuilder,
+}
+
+impl PGCDatabaseBuilder {
+    fn new(
+        schema: SchemaRef,
+        catalog_name: String,
+        catalog_manager: Weak<dyn CatalogManager>,
+        namespace_oid_map: PGNamespaceOidMapRef,
+    ) -> Self {
+        Self {
+            schema,
+            catalog_name,
+            catalog_manager,
+            namespace_oid_map,
+
+            oid: UInt32VectorBuilder::with_capacity(INIT_CAPACITY),
+            datname: StringVectorBuilder::with_capacity(INIT_CAPACITY),
+        }
+    }
+
+    async fn make_database(&mut self, request: Option<ScanRequest>) -> Result<RecordBatch> {
+        let catalog_name = self.catalog_name.clone();
+        let catalog_manager = self
+            .catalog_manager
+            .upgrade()
+            .context(UpgradeWeakCatalogManagerRefSnafu)?;
+        let predicates = Predicates::from_scan_request(&request);
+        for schema_name in catalog_manager
+            .schema_names(&catalog_name, query_ctx())
+            .await?
+        {
+            self.add_database(&predicates, &schema_name);
+        }
+        self.finish()
+    }
+
+    fn add_database(&mut self, predicates: &Predicates, schema_name: &str) {
+        let oid = self.namespace_oid_map.get_oid(schema_name);
+        let row: [(&str, &Value); 2] = [
+            (OID_COLUMN_NAME, &Value::from(oid)),
+            (DATNAME, &Value::from(schema_name)),
+        ];
+
+        if !predicates.eval(&row) {
+            return;
+        }
+
+        self.oid.push(Some(oid));
+        self.datname.push(Some(schema_name));
+    }
+
+    fn finish(&mut self) -> Result<RecordBatch> {
+        let columns: Vec<VectorRef> =
+            vec![Arc::new(self.oid.finish()), Arc::new(self.datname.finish())];
+        RecordBatch::new(self.schema.clone(), columns).context(CreateRecordBatchSnafu)
+    }
+}

src/catalog/src/system_schema/pg_catalog/pg_namespace.rs

@@ -17,6 +17,7 @@
 
 pub(super) mod oid_map;
 
+use std::fmt;
 use std::sync::{Arc, Weak};
 
 use arrow_schema::SchemaRef as ArrowSchemaRef;
@@ -87,6 +88,15 @@ impl PGNamespace {
     }
 }
 
+impl fmt::Debug for PGNamespace {
+    fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
+        f.debug_struct("PGNamespace")
+            .field("schema", &self.schema)
+            .field("catalog_name", &self.catalog_name)
+            .finish()
+    }
+}
+
 impl SystemTable for PGNamespace {
     fn schema(&self) -> SchemaRef {
         self.schema.clone()

src/catalog/src/system_schema/pg_catalog/table_names.rs

@@ -12,7 +12,11 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
-pub const PG_DATABASE: &str = "pg_databases";
+// https://www.postgresql.org/docs/current/catalog-pg-database.html
+pub const PG_DATABASE: &str = "pg_database";
+// https://www.postgresql.org/docs/current/catalog-pg-namespace.html
 pub const PG_NAMESPACE: &str = "pg_namespace";
+// https://www.postgresql.org/docs/current/catalog-pg-class.html
 pub const PG_CLASS: &str = "pg_class";
+// https://www.postgresql.org/docs/current/catalog-pg-type.html
 pub const PG_TYPE: &str = "pg_type";

src/catalog/src/table_source.rs

@@ -365,7 +365,7 @@ mod tests {
 Projection: person.id AS a, person.name AS b
   Filter: person.id > Int32(500)
     TableScan: person"#,
-            format!("\n{:?}", source.get_logical_plan().unwrap())
+            format!("\n{}", source.get_logical_plan().unwrap())
         );
     }
 }

src/catalog/src/table_source/dummy_catalog.rs

@@ -15,12 +15,12 @@
 //! Dummy catalog for region server.
 
 use std::any::Any;
+use std::fmt;
 use std::sync::Arc;
 
 use async_trait::async_trait;
 use common_catalog::format_full_table_name;
-use datafusion::catalog::schema::SchemaProvider;
-use datafusion::catalog::{CatalogProvider, CatalogProviderList};
+use datafusion::catalog::{CatalogProvider, CatalogProviderList, SchemaProvider};
 use datafusion::datasource::TableProvider;
 use snafu::OptionExt;
 use table::table::adapter::DfTableProviderAdapter;
@@ -41,6 +41,12 @@ impl DummyCatalogList {
     }
 }
 
+impl fmt::Debug for DummyCatalogList {
+    fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
+        f.debug_struct("DummyCatalogList").finish()
+    }
+}
+
 impl CatalogProviderList for DummyCatalogList {
     fn as_any(&self) -> &dyn Any {
         self
@@ -91,6 +97,14 @@ impl CatalogProvider for DummyCatalogProvider {
     }
 }
 
+impl fmt::Debug for DummyCatalogProvider {
+    fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
+        f.debug_struct("DummyCatalogProvider")
+            .field("catalog_name", &self.catalog_name)
+            .finish()
+    }
+}
+
 /// A dummy schema provider for [DummyCatalogList].
 #[derive(Clone)]
 struct DummySchemaProvider {
@@ -127,3 +141,12 @@ impl SchemaProvider for DummySchemaProvider {
         true
     }
 }
+
+impl fmt::Debug for DummySchemaProvider {
+    fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
+        f.debug_struct("DummySchemaProvider")
+            .field("catalog_name", &self.catalog_name)
+            .field("schema_name", &self.schema_name)
+            .finish()
+    }
+}

src/cli/Cargo.toml

@@ -4,6 +4,9 @@ version.workspace = true
 edition.workspace = true
 license.workspace = true
 
+[features]
+pg_kvbackend = ["common-meta/pg_kvbackend"]
+
 [lints]
 workspace = true
 
@@ -15,7 +18,7 @@ cache.workspace = true
 catalog.workspace = true
 chrono.workspace = true
 clap.workspace = true
-client.workspace = true
+client = { workspace = true, features = ["testing"] }
 common-base.workspace = true
 common-catalog.workspace = true
 common-config.workspace = true
@@ -56,8 +59,6 @@ tokio.workspace = true
 tracing-appender.workspace = true
 
 [dev-dependencies]
-client = { workspace = true, features = ["testing"] }
-common-test-util.workspace = true
 common-version.workspace = true
 serde.workspace = true
 tempfile.workspace = true

src/cli/src/bench.rs

@@ -22,6 +22,9 @@ use clap::Parser;
 use common_error::ext::BoxedError;
 use common_meta::key::{TableMetadataManager, TableMetadataManagerRef};
 use common_meta::kv_backend::etcd::EtcdStore;
+use common_meta::kv_backend::memory::MemoryKvBackend;
+#[cfg(feature = "pg_kvbackend")]
+use common_meta::kv_backend::postgres::PgStore;
 use common_meta::peer::Peer;
 use common_meta::rpc::router::{Region, RegionRoute};
 use common_telemetry::info;
@@ -55,18 +58,34 @@ where
 #[derive(Debug, Default, Parser)]
 pub struct BenchTableMetadataCommand {
     #[clap(long)]
-    etcd_addr: String,
+    etcd_addr: Option<String>,
+    #[cfg(feature = "pg_kvbackend")]
+    #[clap(long)]
+    postgres_addr: Option<String>,
     #[clap(long)]
     count: u32,
 }
 
 impl BenchTableMetadataCommand {
     pub async fn build(&self) -> std::result::Result<Box<dyn Tool>, BoxedError> {
-        let etcd_store = EtcdStore::with_endpoints([&self.etcd_addr], 128)
-            .await
-            .unwrap();
+        let kv_backend = if let Some(etcd_addr) = &self.etcd_addr {
+            info!("Using etcd as kv backend");
+            EtcdStore::with_endpoints([etcd_addr], 128).await.unwrap()
+        } else {
+            Arc::new(MemoryKvBackend::new())
+        };
 
-        let table_metadata_manager = Arc::new(TableMetadataManager::new(etcd_store));
+        #[cfg(feature = "pg_kvbackend")]
+        let kv_backend = if let Some(postgres_addr) = &self.postgres_addr {
+            info!("Using postgres as kv backend");
+            PgStore::with_url(postgres_addr, "greptime_metakv", 128)
+                .await
+                .unwrap()
+        } else {
+            kv_backend
+        };
+
+        let table_metadata_manager = Arc::new(TableMetadataManager::new(kv_backend));
 
         let tool = BenchTableMetadata {
             table_metadata_manager,

src/cmd/Cargo.toml

@@ -10,9 +10,8 @@ name = "greptime"
 path = "src/bin/greptime.rs"
 
 [features]
-default = ["python", "servers/pprof", "servers/mem-prof"]
+default = ["servers/pprof", "servers/mem-prof"]
 tokio-console = ["common-telemetry/tokio-console"]
-python = ["frontend/python"]
 
 [lints]
 workspace = true
@@ -58,6 +57,7 @@ humantime.workspace = true
 lazy_static.workspace = true
 meta-client.workspace = true
 meta-srv.workspace = true
+metric-engine.workspace = true
 mito2.workspace = true
 moka.workspace = true
 nu-ansi-term = "0.46"

src/cmd/src/cli.rs

@@ -51,8 +51,7 @@ impl App for Instance {
     }
 
     async fn start(&mut self) -> Result<()> {
-        self.start().await.unwrap();
-        Ok(())
+        self.start().await
     }
 
     fn wait_signal(&self) -> bool {

src/cmd/src/datanode.rs

@@ -62,6 +62,11 @@ impl Instance {
     pub fn datanode(&self) -> &Datanode {
         &self.datanode
     }
+
+    /// allow customizing datanode for downstream projects
+    pub fn datanode_mut(&mut self) -> &mut Datanode {
+        &mut self.datanode
+    }
 }
 
 #[async_trait]
@@ -271,7 +276,8 @@ impl StartCommand {
         info!("Datanode options: {:#?}", opts);
 
         let plugin_opts = opts.plugins;
-        let opts = opts.component;
+        let mut opts = opts.component;
+        opts.grpc.detect_hostname();
         let mut plugins = Plugins::new();
         plugins::setup_datanode_plugins(&mut plugins, &plugin_opts, &opts)
             .await

src/cmd/src/error.rs

@@ -345,6 +345,13 @@ pub enum Error {
         #[snafu(implicit)]
         location: Location,
     },
+
+    #[snafu(display("Failed to build wal options allocator"))]
+    BuildWalOptionsAllocator {
+        #[snafu(implicit)]
+        location: Location,
+        source: common_meta::error::Error,
+    },
 }
 
 pub type Result<T> = std::result::Result<T, Error>;
@@ -378,7 +385,8 @@ impl ErrorExt for Error {
 
             Error::StartProcedureManager { source, .. }
             | Error::StopProcedureManager { source, .. } => source.status_code(),
-            Error::StartWalOptionsAllocator { source, .. } => source.status_code(),
+            Error::BuildWalOptionsAllocator { source, .. }
+            | Error::StartWalOptionsAllocator { source, .. } => source.status_code(),
             Error::ReplCreation { .. } | Error::Readline { .. } | Error::HttpQuerySql { .. } => {
                 StatusCode::Internal
             }

src/cmd/src/flownode.rs

@@ -13,6 +13,7 @@
 // limitations under the License.
 
 use std::sync::Arc;
+use std::time::Duration;
 
 use cache::{build_fundamental_cache_registry, with_default_composite_cache_registry};
 use catalog::information_extension::DistributedInformationExtension;
@@ -66,6 +67,11 @@ impl Instance {
     pub fn flownode(&self) -> &FlownodeInstance {
         &self.flownode
     }
+
+    /// allow customizing flownode for downstream projects
+    pub fn flownode_mut(&mut self) -> &mut FlownodeInstance {
+        &mut self.flownode
+    }
 }
 
 #[async_trait::async_trait]
@@ -137,6 +143,11 @@ struct StartCommand {
     /// The prefix of environment variables, default is `GREPTIMEDB_FLOWNODE`;
     #[clap(long, default_value = "GREPTIMEDB_FLOWNODE")]
     env_prefix: String,
+    #[clap(long)]
+    http_addr: Option<String>,
+    /// HTTP request timeout in seconds.
+    #[clap(long)]
+    http_timeout: Option<u64>,
 }
 
 impl StartCommand {
@@ -193,6 +204,14 @@ impl StartCommand {
             opts.mode = Mode::Distributed;
         }
 
+        if let Some(http_addr) = &self.http_addr {
+            opts.http.addr.clone_from(http_addr);
+        }
+
+        if let Some(http_timeout) = self.http_timeout {
+            opts.http.timeout = Duration::from_secs(http_timeout);
+        }
+
         if let (Mode::Distributed, None) = (&opts.mode, &opts.node_id) {
             return MissingConfigSnafu {
                 msg: "Missing node id option",
@@ -217,7 +236,8 @@ impl StartCommand {
         info!("Flownode start command: {:#?}", self);
         info!("Flownode options: {:#?}", opts);
 
-        let opts = opts.component;
+        let mut opts = opts.component;
+        opts.grpc.detect_hostname();
 
         // TODO(discord9): make it not optionale after cluster id is required
         let cluster_id = opts.cluster_id.unwrap_or(0);

src/cmd/src/frontend.rs

@@ -268,7 +268,8 @@ impl StartCommand {
         info!("Frontend options: {:#?}", opts);
 
         let plugin_opts = opts.plugins;
-        let opts = opts.component;
+        let mut opts = opts.component;
+        opts.grpc.detect_hostname();
         let mut plugins = Plugins::new();
         plugins::setup_frontend_plugins(&mut plugins, &plugin_opts, &opts)
             .await

src/cmd/src/metasrv.rs

@@ -249,8 +249,6 @@ impl StartCommand {
 
         if let Some(backend) = &self.backend {
             opts.backend.clone_from(backend);
-        } else {
-            opts.backend = BackendImpl::default()
         }
 
         // Disable dashboard in metasrv.
@@ -274,7 +272,8 @@ impl StartCommand {
         info!("Metasrv options: {:#?}", opts);
 
         let plugin_opts = opts.plugins;
-        let opts = opts.component;
+        let mut opts = opts.component;
+        opts.detect_server_addr();
         let mut plugins = Plugins::new();
         plugins::setup_metasrv_plugins(&mut plugins, &plugin_opts, &opts)
             .await

src/cmd/src/standalone.rs

@@ -22,6 +22,7 @@ use catalog::information_schema::InformationExtension;
 use catalog::kvbackend::KvBackendCatalogManager;
 use clap::Parser;
 use client::api::v1::meta::RegionRole;
+use common_base::readable_size::ReadableSize;
 use common_base::Plugins;
 use common_catalog::consts::{MIN_USER_FLOW_ID, MIN_USER_TABLE_ID};
 use common_config::{metadata_store_dir, Configurable, KvBackendConfig};
@@ -42,7 +43,7 @@ use common_meta::node_manager::NodeManagerRef;
 use common_meta::peer::Peer;
 use common_meta::region_keeper::MemoryRegionKeeper;
 use common_meta::sequence::SequenceBuilder;
-use common_meta::wal_options_allocator::{WalOptionsAllocator, WalOptionsAllocatorRef};
+use common_meta::wal_options_allocator::{build_wal_options_allocator, WalOptionsAllocatorRef};
 use common_procedure::{ProcedureInfo, ProcedureManagerRef};
 use common_telemetry::info;
 use common_telemetry::logging::{LoggingOptions, TracingOptions};
@@ -53,7 +54,7 @@ use datanode::config::{DatanodeOptions, ProcedureConfig, RegionEngineConfig, Sto
 use datanode::datanode::{Datanode, DatanodeBuilder};
 use datanode::region_server::RegionServer;
 use file_engine::config::EngineConfig as FileEngineConfig;
-use flow::{FlowWorkerManager, FlownodeBuilder, FrontendInvoker};
+use flow::{FlowConfig, FlowWorkerManager, FlownodeBuilder, FlownodeOptions, FrontendInvoker};
 use frontend::frontend::FrontendOptions;
 use frontend::instance::builder::FrontendBuilder;
 use frontend::instance::{FrontendInstance, Instance as FeInstance, StandaloneDatanodeManager};
@@ -75,10 +76,10 @@ use tokio::sync::{broadcast, RwLock};
 use tracing_appender::non_blocking::WorkerGuard;
 
 use crate::error::{
-    BuildCacheRegistrySnafu, CreateDirSnafu, IllegalConfigSnafu, InitDdlManagerSnafu,
-    InitMetadataSnafu, InitTimezoneSnafu, LoadLayeredConfigSnafu, OtherSnafu, Result,
-    ShutdownDatanodeSnafu, ShutdownFlownodeSnafu, ShutdownFrontendSnafu, StartDatanodeSnafu,
-    StartFlownodeSnafu, StartFrontendSnafu, StartProcedureManagerSnafu,
+    BuildCacheRegistrySnafu, BuildWalOptionsAllocatorSnafu, CreateDirSnafu, IllegalConfigSnafu,
+    InitDdlManagerSnafu, InitMetadataSnafu, InitTimezoneSnafu, LoadLayeredConfigSnafu, OtherSnafu,
+    Result, ShutdownDatanodeSnafu, ShutdownFlownodeSnafu, ShutdownFrontendSnafu,
+    StartDatanodeSnafu, StartFlownodeSnafu, StartFrontendSnafu, StartProcedureManagerSnafu,
     StartWalOptionsAllocatorSnafu, StopProcedureManagerSnafu,
 };
 use crate::options::{GlobalOptions, GreptimeOptions};
@@ -144,6 +145,7 @@ pub struct StandaloneOptions {
     pub storage: StorageConfig,
     pub metadata_store: KvBackendConfig,
     pub procedure: ProcedureConfig,
+    pub flow: FlowConfig,
     pub logging: LoggingOptions,
     pub user_provider: Option<String>,
     /// Options for different store engines.
@@ -152,6 +154,7 @@ pub struct StandaloneOptions {
     pub tracing: TracingOptions,
     pub init_regions_in_background: bool,
     pub init_regions_parallelism: usize,
+    pub max_in_flight_write_bytes: Option<ReadableSize>,
 }
 
 impl Default for StandaloneOptions {
@@ -171,6 +174,7 @@ impl Default for StandaloneOptions {
             storage: StorageConfig::default(),
             metadata_store: KvBackendConfig::default(),
             procedure: ProcedureConfig::default(),
+            flow: FlowConfig::default(),
             logging: LoggingOptions::default(),
             export_metrics: ExportMetricsOption::default(),
             user_provider: None,
@@ -181,6 +185,7 @@ impl Default for StandaloneOptions {
             tracing: TracingOptions::default(),
             init_regions_in_background: false,
             init_regions_parallelism: 16,
+            max_in_flight_write_bytes: None,
         }
     }
 }
@@ -218,6 +223,7 @@ impl StandaloneOptions {
             user_provider: cloned_opts.user_provider,
             // Handle the export metrics task run by standalone to frontend for execution
             export_metrics: cloned_opts.export_metrics,
+            max_in_flight_write_bytes: cloned_opts.max_in_flight_write_bytes,
             ..Default::default()
         }
     }
@@ -457,7 +463,8 @@ impl StartCommand {
 
         let mut plugins = Plugins::new();
         let plugin_opts = opts.plugins;
-        let opts = opts.component;
+        let mut opts = opts.component;
+        opts.grpc.detect_hostname();
         let fe_opts = opts.frontend_options();
         let dn_opts = opts.datanode_options();
 
@@ -518,8 +525,12 @@ impl StartCommand {
             Self::create_table_metadata_manager(kv_backend.clone()).await?;
 
         let flow_metadata_manager = Arc::new(FlowMetadataManager::new(kv_backend.clone()));
+        let flownode_options = FlownodeOptions {
+            flow: opts.flow.clone(),
+            ..Default::default()
+        };
         let flow_builder = FlownodeBuilder::new(
-            Default::default(),
+            flownode_options,
             plugins.clone(),
             table_metadata_manager.clone(),
             catalog_manager.clone(),
@@ -558,10 +569,11 @@ impl StartCommand {
                 .step(10)
                 .build(),
         );
-        let wal_options_allocator = Arc::new(WalOptionsAllocator::new(
-            opts.wal.clone().into(),
-            kv_backend.clone(),
-        ));
+        let kafka_options = opts.wal.clone().into();
+        let wal_options_allocator = build_wal_options_allocator(&kafka_options, kv_backend.clone())
+            .await
+            .context(BuildWalOptionsAllocatorSnafu)?;
+        let wal_options_allocator = Arc::new(wal_options_allocator);
         let table_meta_allocator = Arc::new(TableMetadataAllocator::new(
             table_id_sequence,
             wal_options_allocator.clone(),

src/cmd/tests/load_config_test.rs

@@ -25,14 +25,16 @@ use common_telemetry::logging::{LoggingOptions, SlowQueryOptions, DEFAULT_OTLP_E
 use common_wal::config::raft_engine::RaftEngineConfig;
 use common_wal::config::DatanodeWalConfig;
 use datanode::config::{DatanodeOptions, RegionEngineConfig, StorageConfig};
-use file_engine::config::EngineConfig;
+use file_engine::config::EngineConfig as FileEngineConfig;
 use frontend::frontend::FrontendOptions;
 use meta_client::MetaClientOptions;
 use meta_srv::metasrv::MetasrvOptions;
 use meta_srv::selector::SelectorType;
+use metric_engine::config::EngineConfig as MetricEngineConfig;
 use mito2::config::MitoConfig;
 use servers::export_metrics::ExportMetricsOption;
 use servers::grpc::GrpcOptions;
+use servers::http::HttpOptions;
 
 #[allow(deprecated)]
 #[test]
@@ -69,10 +71,13 @@ fn test_load_datanode_example_config() {
             region_engine: vec![
                 RegionEngineConfig::Mito(MitoConfig {
                     auto_flush_interval: Duration::from_secs(3600),
-                    experimental_write_cache_ttl: Some(Duration::from_secs(60 * 60 * 8)),
+                    write_cache_ttl: Some(Duration::from_secs(60 * 60 * 8)),
                     ..Default::default()
                 }),
-                RegionEngineConfig::File(EngineConfig {}),
+                RegionEngineConfig::File(FileEngineConfig {}),
+                RegionEngineConfig::Metric(MetricEngineConfig {
+                    experimental_sparse_primary_key_encoding: false,
+                }),
             ],
             logging: LoggingOptions {
                 level: Some("info".to_string()),
@@ -85,7 +90,9 @@ fn test_load_datanode_example_config() {
                 remote_write: Some(Default::default()),
                 ..Default::default()
             },
-            grpc: GrpcOptions::default().with_addr("127.0.0.1:3001"),
+            grpc: GrpcOptions::default()
+                .with_addr("127.0.0.1:3001")
+                .with_hostname("127.0.0.1:3001"),
             rpc_addr: Some("127.0.0.1:3001".to_string()),
             rpc_hostname: Some("127.0.0.1".to_string()),
             rpc_runtime_size: Some(8),
@@ -137,6 +144,11 @@ fn test_load_frontend_example_config() {
                 remote_write: Some(Default::default()),
                 ..Default::default()
             },
+            grpc: GrpcOptions::default().with_hostname("127.0.0.1:4001"),
+            http: HttpOptions {
+                cors_allowed_origins: vec!["https://example.com".to_string()],
+                ..Default::default()
+            },
             ..Default::default()
         },
         ..Default::default()
@@ -154,6 +166,7 @@ fn test_load_metasrv_example_config() {
         component: MetasrvOptions {
             selector: SelectorType::default(),
             data_home: "/tmp/metasrv/".to_string(),
+            server_addr: "127.0.0.1:3002".to_string(),
             logging: LoggingOptions {
                 dir: "/tmp/greptimedb/logs".to_string(),
                 level: Some("info".to_string()),
@@ -203,10 +216,13 @@ fn test_load_standalone_example_config() {
             region_engine: vec![
                 RegionEngineConfig::Mito(MitoConfig {
                     auto_flush_interval: Duration::from_secs(3600),
-                    experimental_write_cache_ttl: Some(Duration::from_secs(60 * 60 * 8)),
+                    write_cache_ttl: Some(Duration::from_secs(60 * 60 * 8)),
                     ..Default::default()
                 }),
-                RegionEngineConfig::File(EngineConfig {}),
+                RegionEngineConfig::File(FileEngineConfig {}),
+                RegionEngineConfig::Metric(MetricEngineConfig {
+                    experimental_sparse_primary_key_encoding: false,
+                }),
             ],
             storage: StorageConfig {
                 data_home: "/tmp/greptimedb/".to_string(),
@@ -223,6 +239,10 @@ fn test_load_standalone_example_config() {
                 remote_write: Some(Default::default()),
                 ..Default::default()
             },
+            http: HttpOptions {
+                cors_allowed_origins: vec!["https://example.com".to_string()],
+                ..Default::default()
+            },
             ..Default::default()
         },
         ..Default::default()

src/common/base/Cargo.toml

@@ -4,6 +4,9 @@ version.workspace = true
 edition.workspace = true
 license.workspace = true
 
+[features]
+testing = []
+
 [lints]
 workspace = true
 

src/common/base/src/range_read.rs

@@ -17,6 +17,7 @@ use std::io;
 use std::ops::Range;
 use std::path::Path;
 use std::pin::Pin;
+use std::sync::atomic::{AtomicU64, Ordering};
 use std::sync::Arc;
 use std::task::{Context, Poll};
 
@@ -33,19 +34,22 @@ pub struct Metadata {
     pub content_length: u64,
 }
 
-/// `RangeReader` reads a range of bytes from a source.
-#[async_trait]
-pub trait RangeReader: Send + Unpin {
+/// `SizeAwareRangeReader` is a `RangeReader` that supports setting a file size hint.
+pub trait SizeAwareRangeReader: RangeReader {
     /// Sets the file size hint for the reader.
     ///
     /// It's used to optimize the reading process by reducing the number of remote requests.
     fn with_file_size_hint(&mut self, file_size_hint: u64);
+}
 
+/// `RangeReader` reads a range of bytes from a source.
+#[async_trait]
+pub trait RangeReader: Sync + Send + Unpin {
     /// Returns the metadata of the source.
-    async fn metadata(&mut self) -> io::Result<Metadata>;
+    async fn metadata(&self) -> io::Result<Metadata>;
 
     /// Reads the bytes in the given range.
-    async fn read(&mut self, range: Range<u64>) -> io::Result<Bytes>;
+    async fn read(&self, range: Range<u64>) -> io::Result<Bytes>;
 
     /// Reads the bytes in the given range into the buffer.
     ///
@@ -53,18 +57,14 @@ pub trait RangeReader: Send + Unpin {
     /// - If the buffer is insufficient to hold the bytes, it will either:
     ///   - Allocate additional space (e.g., for `Vec<u8>`)
     ///   - Panic (e.g., for `&mut [u8]`)
-    async fn read_into(
-        &mut self,
-        range: Range<u64>,
-        buf: &mut (impl BufMut + Send),
-    ) -> io::Result<()> {
+    async fn read_into(&self, range: Range<u64>, buf: &mut (impl BufMut + Send)) -> io::Result<()> {
         let bytes = self.read(range).await?;
         buf.put_slice(&bytes);
         Ok(())
     }
 
     /// Reads the bytes in the given ranges.
-    async fn read_vec(&mut self, ranges: &[Range<u64>]) -> io::Result<Vec<Bytes>> {
+    async fn read_vec(&self, ranges: &[Range<u64>]) -> io::Result<Vec<Bytes>> {
         let mut result = Vec::with_capacity(ranges.len());
         for range in ranges {
             result.push(self.read(range.clone()).await?);
@@ -74,25 +74,20 @@ pub trait RangeReader: Send + Unpin {
 }
 
 #[async_trait]
-impl<R: ?Sized + RangeReader> RangeReader for &mut R {
-    fn with_file_size_hint(&mut self, file_size_hint: u64) {
-        (*self).with_file_size_hint(file_size_hint)
-    }
-
-    async fn metadata(&mut self) -> io::Result<Metadata> {
+impl<R: ?Sized + RangeReader> RangeReader for &R {
+    async fn metadata(&self) -> io::Result<Metadata> {
         (*self).metadata().await
     }
-    async fn read(&mut self, range: Range<u64>) -> io::Result<Bytes> {
+
+    async fn read(&self, range: Range<u64>) -> io::Result<Bytes> {
         (*self).read(range).await
     }
-    async fn read_into(
-        &mut self,
-        range: Range<u64>,
-        buf: &mut (impl BufMut + Send),
-    ) -> io::Result<()> {
+
+    async fn read_into(&self, range: Range<u64>, buf: &mut (impl BufMut + Send)) -> io::Result<()> {
         (*self).read_into(range, buf).await
     }
-    async fn read_vec(&mut self, ranges: &[Range<u64>]) -> io::Result<Vec<Bytes>> {
+
+    async fn read_vec(&self, ranges: &[Range<u64>]) -> io::Result<Vec<Bytes>> {
         (*self).read_vec(ranges).await
     }
 }
@@ -120,7 +115,7 @@ pub struct AsyncReadAdapter<R> {
 
 impl<R: RangeReader + 'static> AsyncReadAdapter<R> {
     pub async fn new(inner: R) -> io::Result<Self> {
-        let mut inner = inner;
+        let inner = inner;
         let metadata = inner.metadata().await?;
         Ok(AsyncReadAdapter {
             inner: Arc::new(Mutex::new(inner)),
@@ -160,7 +155,7 @@ impl<R: RangeReader + 'static> AsyncRead for AsyncReadAdapter<R> {
             let range = *this.position..(*this.position + size);
             let inner = this.inner.clone();
             let fut = async move {
-                let mut inner = inner.lock().await;
+                let inner = inner.lock().await;
                 inner.read(range).await
             };
 
@@ -195,27 +190,24 @@ impl<R: RangeReader + 'static> AsyncRead for AsyncReadAdapter<R> {
 
 #[async_trait]
 impl RangeReader for Vec<u8> {
-    fn with_file_size_hint(&mut self, _file_size_hint: u64) {
-        // do nothing
-    }
-
-    async fn metadata(&mut self) -> io::Result<Metadata> {
+    async fn metadata(&self) -> io::Result<Metadata> {
         Ok(Metadata {
             content_length: self.len() as u64,
         })
     }
 
-    async fn read(&mut self, range: Range<u64>) -> io::Result<Bytes> {
+    async fn read(&self, range: Range<u64>) -> io::Result<Bytes> {
         let bytes = Bytes::copy_from_slice(&self[range.start as usize..range.end as usize]);
         Ok(bytes)
     }
 }
 
+// TODO(weny): considers replacing `tokio::fs::File` with opendal reader.
 /// `FileReader` is a `RangeReader` for reading a file.
 pub struct FileReader {
     content_length: u64,
-    position: u64,
-    file: tokio::fs::File,
+    position: AtomicU64,
+    file: Mutex<tokio::fs::File>,
 }
 
 impl FileReader {
@@ -225,32 +217,36 @@ impl FileReader {
         let metadata = file.metadata().await?;
         Ok(FileReader {
             content_length: metadata.len(),
-            position: 0,
-            file,
+            position: AtomicU64::new(0),
+            file: Mutex::new(file),
         })
     }
 }
 
+impl SizeAwareRangeReader for FileReader {
+    fn with_file_size_hint(&mut self, _file_size_hint: u64) {
+        // do nothing
+    }
+}
+
 #[async_trait]
 impl RangeReader for FileReader {
-    fn with_file_size_hint(&mut self, _file_size_hint: u64) {
-        // do nothing
-    }
-
-    async fn metadata(&mut self) -> io::Result<Metadata> {
+    async fn metadata(&self) -> io::Result<Metadata> {
         Ok(Metadata {
             content_length: self.content_length,
         })
     }
 
-    async fn read(&mut self, mut range: Range<u64>) -> io::Result<Bytes> {
-        if range.start != self.position {
-            self.file.seek(io::SeekFrom::Start(range.start)).await?;
-            self.position = range.start;
+    async fn read(&self, mut range: Range<u64>) -> io::Result<Bytes> {
+        let mut file = self.file.lock().await;
+
+        if range.start != self.position.load(Ordering::Relaxed) {
+            file.seek(io::SeekFrom::Start(range.start)).await?;
+            self.position.store(range.start, Ordering::Relaxed);
         }
 
         range.end = range.end.min(self.content_length);
-        if range.end <= self.position {
+        if range.end <= self.position.load(Ordering::Relaxed) {
             return Err(io::Error::new(
                 io::ErrorKind::UnexpectedEof,
                 "Start of range is out of bounds",
@@ -259,8 +255,8 @@ impl RangeReader for FileReader {
 
         let mut buf = vec![0; (range.end - range.start) as usize];
 
-        self.file.read_exact(&mut buf).await?;
-        self.position = range.end;
+        file.read_exact(&mut buf).await?;
+        self.position.store(range.end, Ordering::Relaxed);
 
         Ok(Bytes::from(buf))
     }
@@ -301,7 +297,7 @@ mod tests {
         let data = b"hello world";
         tokio::fs::write(path, data).await.unwrap();
 
-        let mut reader = FileReader::new(path).await.unwrap();
+        let reader = FileReader::new(path).await.unwrap();
         let metadata = reader.metadata().await.unwrap();
         assert_eq!(metadata.content_length, data.len() as u64);
 

src/common/catalog/src/consts.rs

@@ -109,6 +109,7 @@ pub const INFORMATION_SCHEMA_REGION_STATISTICS_TABLE_ID: u32 = 35;
 pub const PG_CATALOG_PG_CLASS_TABLE_ID: u32 = 256;
 pub const PG_CATALOG_PG_TYPE_TABLE_ID: u32 = 257;
 pub const PG_CATALOG_PG_NAMESPACE_TABLE_ID: u32 = 258;
+pub const PG_CATALOG_PG_DATABASE_TABLE_ID: u32 = 259;
 
 // ----- End of pg_catalog tables -----
 

src/common/config/src/config.rs

@@ -73,14 +73,21 @@ pub trait Configurable: Serialize + DeserializeOwned + Default + Sized {
             layered_config = layered_config.add_source(File::new(config_file, FileFormat::Toml));
         }
 
-        let opts = layered_config
+        let mut opts: Self = layered_config
             .build()
             .and_then(|x| x.try_deserialize())
             .context(LoadLayeredConfigSnafu)?;
 
+        opts.validate_sanitize()?;
+
         Ok(opts)
     }
 
+    /// Validate(and possibly sanitize) the configuration.
+    fn validate_sanitize(&mut self) -> Result<()> {
+        Ok(())
+    }
+
     /// List of toml keys that should be parsed as a list.
     fn env_list_keys() -> Option<&'static [&'static str]> {
         None

src/common/datasource/Cargo.toml

@@ -31,7 +31,7 @@ derive_builder.workspace = true
 futures.workspace = true
 lazy_static.workspace = true
 object-store.workspace = true
-orc-rust = { git = "https://github.com/datafusion-contrib/datafusion-orc.git", rev = "502217315726314c4008808fe169764529640599", default-features = false, features = [
+orc-rust = { version = "0.5", default-features = false, features = [
     "async",
 ] }
 parquet.workspace = true

src/common/datasource/src/error.rs

@@ -180,7 +180,7 @@ pub enum Error {
 
     #[snafu(display("Failed to parse format {} with value: {}", key, value))]
     ParseFormat {
-        key: &'static str,
+        key: String,
         value: String,
         #[snafu(implicit)]
         location: Location,

src/common/datasource/src/file_format.rs

@@ -126,8 +126,7 @@ impl ArrowDecoder for arrow::csv::reader::Decoder {
     }
 }
 
-#[allow(deprecated)]
-impl ArrowDecoder for arrow::json::RawDecoder {
+impl ArrowDecoder for arrow::json::reader::Decoder {
     fn decode(&mut self, buf: &[u8]) -> result::Result<usize, ArrowError> {
         self.decode(buf)
     }

src/common/datasource/src/file_format/csv.rs

@@ -17,8 +17,7 @@ use std::str::FromStr;
 use std::sync::Arc;
 
 use arrow::csv;
-#[allow(deprecated)]
-use arrow::csv::reader::infer_reader_schema as infer_csv_schema;
+use arrow::csv::reader::Format;
 use arrow::record_batch::RecordBatch;
 use arrow_schema::{Schema, SchemaRef};
 use async_trait::async_trait;
@@ -161,7 +160,6 @@ impl FileOpener for CsvOpener {
     }
 }
 
-#[allow(deprecated)]
 #[async_trait]
 impl FileFormat for CsvFormat {
     async fn infer_schema(&self, store: &ObjectStore, path: &str) -> Result<Schema> {
@@ -188,9 +186,12 @@ impl FileFormat for CsvFormat {
         common_runtime::spawn_blocking_global(move || {
             let reader = SyncIoBridge::new(decoded);
 
-            let (schema, _records_read) =
-                infer_csv_schema(reader, delimiter, schema_infer_max_record, has_header)
-                    .context(error::InferSchemaSnafu)?;
+            let format = Format::default()
+                .with_delimiter(delimiter)
+                .with_header(has_header);
+            let (schema, _records_read) = format
+                .infer_schema(reader, schema_infer_max_record)
+                .context(error::InferSchemaSnafu)?;
             Ok(schema)
         })
         .await
@@ -253,7 +254,7 @@ mod tests {
                 "c7: Int64: NULL",
                 "c8: Int64: NULL",
                 "c9: Int64: NULL",
-                "c10: Int64: NULL",
+                "c10: Utf8: NULL",
                 "c11: Float64: NULL",
                 "c12: Float64: NULL",
                 "c13: Utf8: NULL"

src/common/datasource/src/file_format/json.rs

@@ -20,8 +20,7 @@ use std::sync::Arc;
 use arrow::datatypes::SchemaRef;
 use arrow::json::reader::{infer_json_schema_from_iterator, ValueIter};
 use arrow::json::writer::LineDelimited;
-#[allow(deprecated)]
-use arrow::json::{self, RawReaderBuilder};
+use arrow::json::{self, ReaderBuilder};
 use arrow::record_batch::RecordBatch;
 use arrow_schema::Schema;
 use async_trait::async_trait;
@@ -140,7 +139,6 @@ impl JsonOpener {
     }
 }
 
-#[allow(deprecated)]
 impl FileOpener for JsonOpener {
     fn open(&self, meta: FileMeta) -> DataFusionResult<FileOpenFuture> {
         open_with_decoder(
@@ -148,7 +146,7 @@ impl FileOpener for JsonOpener {
             meta.location().to_string(),
             self.compression_type,
             || {
-                RawReaderBuilder::new(self.projected_schema.clone())
+                ReaderBuilder::new(self.projected_schema.clone())
                     .with_batch_size(self.batch_size)
                     .build_decoder()
                     .map_err(DataFusionError::from)

src/common/datasource/src/file_format/tests.rs

@@ -42,7 +42,7 @@ struct Test<'a, T: FileOpener> {
     expected: Vec<&'a str>,
 }
 
-impl<'a, T: FileOpener> Test<'a, T> {
+impl<T: FileOpener> Test<'_, T> {
     pub async fn run(self) {
         let result = FileStream::new(
             &self.config,

src/common/datasource/tests/orc/write.py

@@ -35,10 +35,23 @@ data = {
     "bigint_other": [5, -5, 1, 5, 5],
     "utf8_increase": ["a", "bb", "ccc", "dddd", "eeeee"],
     "utf8_decrease": ["eeeee", "dddd", "ccc", "bb", "a"],
-    "timestamp_simple": [datetime.datetime(2023, 4, 1, 20, 15, 30, 2000), datetime.datetime.fromtimestamp(int('1629617204525777000')/1000000000), datetime.datetime(2023, 1, 1), datetime.datetime(2023, 2, 1), datetime.datetime(2023, 3, 1)],
-    "date_simple": [datetime.date(2023, 4, 1), datetime.date(2023, 3, 1), datetime.date(2023, 1, 1), datetime.date(2023, 2, 1), datetime.date(2023, 3, 1)]
+    "timestamp_simple": [
+        datetime.datetime(2023, 4, 1, 20, 15, 30, 2000),
+        datetime.datetime.fromtimestamp(int("1629617204525777000") / 1000000000),
+        datetime.datetime(2023, 1, 1),
+        datetime.datetime(2023, 2, 1),
+        datetime.datetime(2023, 3, 1),
+    ],
+    "date_simple": [
+        datetime.date(2023, 4, 1),
+        datetime.date(2023, 3, 1),
+        datetime.date(2023, 1, 1),
+        datetime.date(2023, 2, 1),
+        datetime.date(2023, 3, 1),
+    ],
 }
 
+
 def infer_schema(data):
     schema = "struct<"
     for key, value in data.items():
@@ -56,7 +69,7 @@ def infer_schema(data):
         elif key.startswith("date"):
             dt = "date"
         else:
-            print(key,value,dt)
+            print(key, value, dt)
             raise NotImplementedError
         if key.startswith("double"):
             dt = "double"
@@ -68,7 +81,6 @@ def infer_schema(data):
     return schema
 
 
-
 def _write(
     schema: str,
     data,

src/common/decimal/src/decimal128.rs

@@ -164,7 +164,7 @@ impl FromStr for Decimal128 {
     type Err = Error;
 
     fn from_str(s: &str) -> Result<Self, Self::Err> {
-        let len = s.as_bytes().len();
+        let len = s.len();
         if len <= BYTES_TO_OVERFLOW_RUST_DECIMAL {
             let rd = RustDecimal::from_str_exact(s).context(ParseRustDecimalStrSnafu { raw: s })?;
             Ok(Self::from(rd))

src/common/error/Cargo.toml

@@ -8,6 +8,7 @@ license.workspace = true
 workspace = true
 
 [dependencies]
+http.workspace = true
 snafu.workspace = true
 strum.workspace = true
 tonic.workspace = true

src/common/error/src/lib.rs

@@ -18,9 +18,30 @@ pub mod ext;
 pub mod mock;
 pub mod status_code;
 
+use http::{HeaderMap, HeaderValue};
 pub use snafu;
 
 // HACK - these headers are here for shared in gRPC services. For common HTTP headers,
 // please define in `src/servers/src/http/header.rs`.
 pub const GREPTIME_DB_HEADER_ERROR_CODE: &str = "x-greptime-err-code";
 pub const GREPTIME_DB_HEADER_ERROR_MSG: &str = "x-greptime-err-msg";
+
+/// Create a http header map from error code and message.
+/// using `GREPTIME_DB_HEADER_ERROR_CODE` and `GREPTIME_DB_HEADER_ERROR_MSG` as keys.
+pub fn from_err_code_msg_to_header(code: u32, msg: &str) -> HeaderMap {
+    let mut header = HeaderMap::new();
+
+    let msg = HeaderValue::from_str(msg).unwrap_or_else(|_| {
+        HeaderValue::from_bytes(
+            &msg.as_bytes()
+                .iter()
+                .flat_map(|b| std::ascii::escape_default(*b))
+                .collect::<Vec<u8>>(),
+        )
+        .expect("Already escaped string should be valid ascii")
+    });
+
+    header.insert(GREPTIME_DB_HEADER_ERROR_CODE, code.into());
+    header.insert(GREPTIME_DB_HEADER_ERROR_MSG, msg);
+    header
+}

src/common/function/Cargo.toml

@@ -33,7 +33,7 @@ geo-types = { version = "0.7", optional = true }
 geohash = { version = "0.13", optional = true }
 h3o = { version = "0.6", optional = true }
 jsonb.workspace = true
-nalgebra = "0.33"
+nalgebra.workspace = true
 num = "0.4"
 num-traits = "0.2"
 once_cell.workspace = true

src/common/function/src/function_registry.rs

@@ -25,7 +25,6 @@ use crate::scalars::expression::ExpressionFunction;
 use crate::scalars::json::JsonFunction;
 use crate::scalars::matches::MatchesFunction;
 use crate::scalars::math::MathFunction;
-use crate::scalars::numpy::NumpyFunction;
 use crate::scalars::timestamp::TimestampFunction;
 use crate::scalars::vector::VectorFunction;
 use crate::system::SystemFunction;
@@ -103,7 +102,6 @@ pub static FUNCTION_REGISTRY: Lazy<Arc<FunctionRegistry>> = Lazy::new(|| {
 
     // Utility functions
     MathFunction::register(&function_registry);
-    NumpyFunction::register(&function_registry);
     TimestampFunction::register(&function_registry);
     DateFunction::register(&function_registry);
     ExpressionFunction::register(&function_registry);

src/common/function/src/scalars.rs

@@ -20,7 +20,6 @@ pub mod geo;
 pub mod json;
 pub mod matches;
 pub mod math;
-pub mod numpy;
 pub mod vector;
 
 #[cfg(test)]

src/common/function/src/scalars/aggregate.rs

@@ -32,6 +32,8 @@ pub use scipy_stats_norm_cdf::ScipyStatsNormCdfAccumulatorCreator;
 pub use scipy_stats_norm_pdf::ScipyStatsNormPdfAccumulatorCreator;
 
 use crate::function_registry::FunctionRegistry;
+use crate::scalars::vector::product::VectorProductCreator;
+use crate::scalars::vector::sum::VectorSumCreator;
 
 /// A function creates `AggregateFunctionCreator`.
 /// "Aggregator" *is* AggregatorFunction. Since the later one is long, we named an short alias for it.
@@ -91,6 +93,8 @@ impl AggregateFunctions {
         register_aggr_func!("argmin", 1, ArgminAccumulatorCreator);
         register_aggr_func!("scipystatsnormcdf", 2, ScipyStatsNormCdfAccumulatorCreator);
         register_aggr_func!("scipystatsnormpdf", 2, ScipyStatsNormPdfAccumulatorCreator);
+        register_aggr_func!("vec_sum", 1, VectorSumCreator);
+        register_aggr_func!("vec_product", 1, VectorProductCreator);
 
         #[cfg(feature = "geo")]
         register_aggr_func!(

src/common/function/src/scalars/date/date_add.rs

@@ -91,6 +91,7 @@ mod tests {
     use std::sync::Arc;
 
     use common_query::prelude::{TypeSignature, Volatility};
+    use datatypes::arrow::datatypes::IntervalDayTime;
     use datatypes::prelude::ConcreteDataType;
     use datatypes::value::Value;
     use datatypes::vectors::{
@@ -134,7 +135,12 @@ mod tests {
 
         let times = vec![Some(123), None, Some(42), None];
         // Intervals in milliseconds
-        let intervals = vec![1000, 2000, 3000, 1000];
+        let intervals = vec![
+            IntervalDayTime::new(0, 1000),
+            IntervalDayTime::new(0, 2000),
+            IntervalDayTime::new(0, 3000),
+            IntervalDayTime::new(0, 1000),
+        ];
         let results = [Some(124), None, Some(45), None];
 
         let time_vector = TimestampSecondVector::from(times.clone());

src/common/function/src/scalars/date/date_sub.rs

@@ -91,6 +91,7 @@ mod tests {
     use std::sync::Arc;
 
     use common_query::prelude::{TypeSignature, Volatility};
+    use datatypes::arrow::datatypes::IntervalDayTime;
     use datatypes::prelude::ConcreteDataType;
     use datatypes::value::Value;
     use datatypes::vectors::{
@@ -139,7 +140,12 @@ mod tests {
 
         let times = vec![Some(123), None, Some(42), None];
         // Intervals in milliseconds
-        let intervals = vec![1000, 2000, 3000, 1000];
+        let intervals = vec![
+            IntervalDayTime::new(0, 1000),
+            IntervalDayTime::new(0, 2000),
+            IntervalDayTime::new(0, 3000),
+            IntervalDayTime::new(0, 1000),
+        ];
         let results = [Some(122), None, Some(39), None];
 
         let time_vector = TimestampSecondVector::from(times.clone());

src/common/function/src/scalars/matches.rs

@@ -21,10 +21,9 @@ use common_query::error::{
 };
 use datafusion::common::tree_node::{Transformed, TreeNode, TreeNodeIterator, TreeNodeRecursion};
 use datafusion::common::{DFSchema, Result as DfResult};
-use datafusion::execution::context::SessionState;
+use datafusion::execution::SessionStateBuilder;
 use datafusion::logical_expr::{self, Expr, Volatility};
 use datafusion::physical_planner::{DefaultPhysicalPlanner, PhysicalPlanner};
-use datafusion::prelude::SessionConfig;
 use datatypes::arrow::array::RecordBatch;
 use datatypes::arrow::datatypes::{DataType, Field};
 use datatypes::prelude::VectorRef;
@@ -104,8 +103,7 @@ impl MatchesFunction {
         let like_expr = ast.into_like_expr(col_name);
 
         let input_schema = Self::input_schema();
-        let session_state =
-            SessionState::new_with_config_rt(SessionConfig::default(), Arc::default());
+        let session_state = SessionStateBuilder::new().with_default_features().build();
         let planner = DefaultPhysicalPlanner::default();
         let physical_expr = planner
             .create_physical_expr(&like_expr, &input_schema, &session_state)
@@ -131,7 +129,7 @@ impl MatchesFunction {
     }
 
     fn input_schema() -> DFSchema {
-        DFSchema::from_unqualifed_fields(
+        DFSchema::from_unqualified_fields(
             [Arc::new(Field::new("data", DataType::Utf8, true))].into(),
             HashMap::new(),
         )
@@ -725,7 +723,8 @@ struct Tokenizer {
 impl Tokenizer {
     pub fn tokenize(mut self, pattern: &str) -> Result<Vec<Token>> {
         let mut tokens = vec![];
-        while self.cursor < pattern.len() {
+        let char_len = pattern.chars().count();
+        while self.cursor < char_len {
             // TODO: collect pattern into Vec<char> if this tokenizer is bottleneck in the future
             let c = pattern.chars().nth(self.cursor).unwrap();
             match c {
@@ -794,7 +793,8 @@ impl Tokenizer {
         let mut phase = String::new();
         let mut is_quote_present = false;
 
-        while self.cursor < pattern.len() {
+        let char_len = pattern.chars().count();
+        while self.cursor < char_len {
             let mut c = pattern.chars().nth(self.cursor).unwrap();
 
             match c {
@@ -899,6 +899,26 @@ mod test {
                     Phase("c".to_string()),
                 ],
             ),
+            (
+                r#"中文 测试"#,
+                vec![Phase("中文".to_string()), Phase("测试".to_string())],
+            ),
+            (
+                r#"中文 AND 测试"#,
+                vec![Phase("中文".to_string()), And, Phase("测试".to_string())],
+            ),
+            (
+                r#"中文 +测试"#,
+                vec![Phase("中文".to_string()), Must, Phase("测试".to_string())],
+            ),
+            (
+                r#"中文 -测试"#,
+                vec![
+                    Phase("中文".to_string()),
+                    Negative,
+                    Phase("测试".to_string()),
+                ],
+            ),
         ];
 
         for (query, expected) in cases {
@@ -1030,6 +1050,61 @@ mod test {
                     ],
                 },
             ),
+            (
+                r#"中文 测试"#,
+                PatternAst::Binary {
+                    op: BinaryOp::Or,
+                    children: vec![
+                        PatternAst::Literal {
+                            op: UnaryOp::Optional,
+                            pattern: "中文".to_string(),
+                        },
+                        PatternAst::Literal {
+                            op: UnaryOp::Optional,
+                            pattern: "测试".to_string(),
+                        },
+                    ],
+                },
+            ),
+            (
+                r#"中文 AND 测试"#,
+                PatternAst::Binary {
+                    op: BinaryOp::And,
+                    children: vec![
+                        PatternAst::Literal {
+                            op: UnaryOp::Optional,
+                            pattern: "中文".to_string(),
+                        },
+                        PatternAst::Literal {
+                            op: UnaryOp::Optional,
+                            pattern: "测试".to_string(),
+                        },
+                    ],
+                },
+            ),
+            (
+                r#"中文 +测试"#,
+                PatternAst::Literal {
+                    op: UnaryOp::Must,
+                    pattern: "测试".to_string(),
+                },
+            ),
+            (
+                r#"中文 -测试"#,
+                PatternAst::Binary {
+                    op: BinaryOp::And,
+                    children: vec![
+                        PatternAst::Literal {
+                            op: UnaryOp::Negative,
+                            pattern: "测试".to_string(),
+                        },
+                        PatternAst::Literal {
+                            op: UnaryOp::Optional,
+                            pattern: "中文".to_string(),
+                        },
+                    ],
+                },
+            ),
         ];
 
         for (query, expected) in cases {

src/common/function/src/scalars/numpy.rs

@@ -1,30 +0,0 @@
-// Copyright 2023 Greptime Team
-//
-// Licensed under the Apache License, Version 2.0 (the "License");
-// you may not use this file except in compliance with the License.
-// You may obtain a copy of the License at
-//
-//     http://www.apache.org/licenses/LICENSE-2.0
-//
-// Unless required by applicable law or agreed to in writing, software
-// distributed under the License is distributed on an "AS IS" BASIS,
-// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-// See the License for the specific language governing permissions and
-// limitations under the License.
-
-mod clip;
-mod interp;
-
-use std::sync::Arc;
-
-use clip::ClipFunction;
-
-use crate::function_registry::FunctionRegistry;
-
-pub(crate) struct NumpyFunction;
-
-impl NumpyFunction {
-    pub fn register(registry: &FunctionRegistry) {
-        registry.register(Arc::new(ClipFunction));
-    }
-}

src/common/function/src/scalars/numpy/clip.rs

@@ -1,298 +0,0 @@
-// Copyright 2023 Greptime Team
-//
-// Licensed under the Apache License, Version 2.0 (the "License");
-// you may not use this file except in compliance with the License.
-// You may obtain a copy of the License at
-//
-//     http://www.apache.org/licenses/LICENSE-2.0
-//
-// Unless required by applicable law or agreed to in writing, software
-// distributed under the License is distributed on an "AS IS" BASIS,
-// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-// See the License for the specific language governing permissions and
-// limitations under the License.
-
-use std::fmt;
-use std::sync::Arc;
-
-use common_query::error::Result;
-use common_query::prelude::{Signature, Volatility};
-use datatypes::arrow::compute;
-use datatypes::arrow::datatypes::ArrowPrimitiveType;
-use datatypes::data_type::ConcreteDataType;
-use datatypes::prelude::*;
-use datatypes::vectors::PrimitiveVector;
-use paste::paste;
-
-use crate::function::{Function, FunctionContext};
-use crate::scalars::expression::{scalar_binary_op, EvalContext};
-
-/// numpy.clip function, <https://numpy.org/doc/stable/reference/generated/numpy.clip.html>
-#[derive(Clone, Debug, Default)]
-pub struct ClipFunction;
-
-macro_rules! define_eval {
-    ($O: ident) => {
-        paste! {
-            fn [<eval_ $O>](columns: &[VectorRef]) -> Result<VectorRef> {
-                fn cast_vector(input: &VectorRef) -> VectorRef {
-                    Arc::new(PrimitiveVector::<<$O as WrapperType>::LogicalType>::try_from_arrow_array(
-                        compute::cast(&input.to_arrow_array(), &<<<$O as WrapperType>::LogicalType as LogicalPrimitiveType>::ArrowPrimitive as ArrowPrimitiveType>::DATA_TYPE).unwrap()
-                    ).unwrap()) as _
-                }
-                let operator_1 = cast_vector(&columns[0]);
-                let operator_2 = cast_vector(&columns[1]);
-                let operator_3 = cast_vector(&columns[2]);
-
-                // clip(a, min, max) is equals to min(max(a, min), max)
-                let col: VectorRef = Arc::new(scalar_binary_op::<$O, $O, $O, _>(
-                    &operator_1,
-                    &operator_2,
-                    scalar_max,
-                    &mut EvalContext::default(),
-                )?);
-                let col = scalar_binary_op::<$O, $O, $O, _>(
-                    &col,
-                    &operator_3,
-                    scalar_min,
-                    &mut EvalContext::default(),
-                )?;
-                Ok(Arc::new(col))
-            }
-        }
-    };
-}
-
-define_eval!(i64);
-define_eval!(u64);
-define_eval!(f64);
-
-impl Function for ClipFunction {
-    fn name(&self) -> &str {
-        "clip"
-    }
-
-    fn return_type(&self, input_types: &[ConcreteDataType]) -> Result<ConcreteDataType> {
-        if input_types.iter().all(ConcreteDataType::is_signed) {
-            Ok(ConcreteDataType::int64_datatype())
-        } else if input_types.iter().all(ConcreteDataType::is_unsigned) {
-            Ok(ConcreteDataType::uint64_datatype())
-        } else {
-            Ok(ConcreteDataType::float64_datatype())
-        }
-    }
-
-    fn signature(&self) -> Signature {
-        Signature::uniform(3, ConcreteDataType::numerics(), Volatility::Immutable)
-    }
-
-    fn eval(&self, _func_ctx: FunctionContext, columns: &[VectorRef]) -> Result<VectorRef> {
-        if columns.iter().all(|v| v.data_type().is_signed()) {
-            eval_i64(columns)
-        } else if columns.iter().all(|v| v.data_type().is_unsigned()) {
-            eval_u64(columns)
-        } else {
-            eval_f64(columns)
-        }
-    }
-}
-
-#[inline]
-pub fn min<T: PartialOrd>(input: T, min: T) -> T {
-    if input < min {
-        input
-    } else {
-        min
-    }
-}
-
-#[inline]
-pub fn max<T: PartialOrd>(input: T, max: T) -> T {
-    if input > max {
-        input
-    } else {
-        max
-    }
-}
-
-#[inline]
-fn scalar_min<O>(left: Option<O>, right: Option<O>, _ctx: &mut EvalContext) -> Option<O>
-where
-    O: Scalar + Copy + PartialOrd,
-{
-    match (left, right) {
-        (Some(left), Some(right)) => Some(min(left, right)),
-        _ => None,
-    }
-}
-
-#[inline]
-fn scalar_max<O>(left: Option<O>, right: Option<O>, _ctx: &mut EvalContext) -> Option<O>
-where
-    O: Scalar + Copy + PartialOrd,
-{
-    match (left, right) {
-        (Some(left), Some(right)) => Some(max(left, right)),
-        _ => None,
-    }
-}
-
-impl fmt::Display for ClipFunction {
-    fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
-        write!(f, "CLIP")
-    }
-}
-
-#[cfg(test)]
-mod tests {
-    use common_query::prelude::TypeSignature;
-    use datatypes::value::Value;
-    use datatypes::vectors::{
-        ConstantVector, Float32Vector, Int16Vector, Int32Vector, Int8Vector, UInt16Vector,
-        UInt32Vector, UInt8Vector,
-    };
-
-    use super::*;
-
-    #[test]
-    fn test_clip_signature() {
-        let clip = ClipFunction;
-
-        assert_eq!("clip", clip.name());
-        assert_eq!(
-            ConcreteDataType::int64_datatype(),
-            clip.return_type(&[]).unwrap()
-        );
-
-        assert_eq!(
-            ConcreteDataType::int64_datatype(),
-            clip.return_type(&[
-                ConcreteDataType::int16_datatype(),
-                ConcreteDataType::int64_datatype(),
-                ConcreteDataType::int8_datatype()
-            ])
-            .unwrap()
-        );
-        assert_eq!(
-            ConcreteDataType::uint64_datatype(),
-            clip.return_type(&[
-                ConcreteDataType::uint16_datatype(),
-                ConcreteDataType::uint64_datatype(),
-                ConcreteDataType::uint8_datatype()
-            ])
-            .unwrap()
-        );
-        assert_eq!(
-            ConcreteDataType::float64_datatype(),
-            clip.return_type(&[
-                ConcreteDataType::uint16_datatype(),
-                ConcreteDataType::int64_datatype(),
-                ConcreteDataType::uint8_datatype()
-            ])
-            .unwrap()
-        );
-
-        assert!(matches!(clip.signature(),
-                         Signature {
-                             type_signature: TypeSignature::Uniform(3, valid_types),
-                             volatility: Volatility::Immutable
-                         } if  valid_types == ConcreteDataType::numerics()
-        ));
-    }
-
-    #[test]
-    fn test_clip_fn_signed() {
-        // eval with signed integers
-        let args: Vec<VectorRef> = vec![
-            Arc::new(Int32Vector::from_values(0..10)),
-            Arc::new(ConstantVector::new(
-                Arc::new(Int8Vector::from_vec(vec![3])),
-                10,
-            )),
-            Arc::new(ConstantVector::new(
-                Arc::new(Int16Vector::from_vec(vec![6])),
-                10,
-            )),
-        ];
-
-        let vector = ClipFunction
-            .eval(FunctionContext::default(), &args)
-            .unwrap();
-        assert_eq!(10, vector.len());
-
-        // clip([0, 1, 2, 3, 4, 5, 6, 7, 8, 9], 3, 6) = [3, 3, 3, 3, 4, 5, 6, 6, 6, 6]
-        for i in 0..10 {
-            if i <= 3 {
-                assert!(matches!(vector.get(i), Value::Int64(v) if v == 3));
-            } else if i <= 6 {
-                assert!(matches!(vector.get(i), Value::Int64(v) if v == (i as i64)));
-            } else {
-                assert!(matches!(vector.get(i), Value::Int64(v) if v == 6));
-            }
-        }
-    }
-
-    #[test]
-    fn test_clip_fn_unsigned() {
-        // eval with unsigned integers
-        let args: Vec<VectorRef> = vec![
-            Arc::new(UInt8Vector::from_values(0..10)),
-            Arc::new(ConstantVector::new(
-                Arc::new(UInt32Vector::from_vec(vec![3])),
-                10,
-            )),
-            Arc::new(ConstantVector::new(
-                Arc::new(UInt16Vector::from_vec(vec![6])),
-                10,
-            )),
-        ];
-
-        let vector = ClipFunction
-            .eval(FunctionContext::default(), &args)
-            .unwrap();
-        assert_eq!(10, vector.len());
-
-        // clip([0, 1, 2, 3, 4, 5, 6, 7, 8, 9], 3, 6) = [3, 3, 3, 3, 4, 5, 6, 6, 6, 6]
-        for i in 0..10 {
-            if i <= 3 {
-                assert!(matches!(vector.get(i), Value::UInt64(v) if v == 3));
-            } else if i <= 6 {
-                assert!(matches!(vector.get(i), Value::UInt64(v) if v == (i as u64)));
-            } else {
-                assert!(matches!(vector.get(i), Value::UInt64(v) if v == 6));
-            }
-        }
-    }
-
-    #[test]
-    fn test_clip_fn_float() {
-        // eval with floats
-        let args: Vec<VectorRef> = vec![
-            Arc::new(Int8Vector::from_values(0..10)),
-            Arc::new(ConstantVector::new(
-                Arc::new(UInt32Vector::from_vec(vec![3])),
-                10,
-            )),
-            Arc::new(ConstantVector::new(
-                Arc::new(Float32Vector::from_vec(vec![6f32])),
-                10,
-            )),
-        ];
-
-        let vector = ClipFunction
-            .eval(FunctionContext::default(), &args)
-            .unwrap();
-        assert_eq!(10, vector.len());
-
-        // clip([0, 1, 2, 3, 4, 5, 6, 7, 8, 9], 3, 6) = [3, 3, 3, 3, 4, 5, 6, 6, 6, 6]
-        for i in 0..10 {
-            if i <= 3 {
-                assert!(matches!(vector.get(i), Value::Float64(v) if v == 3.0));
-            } else if i <= 6 {
-                assert!(matches!(vector.get(i), Value::Float64(v) if v == (i as f64)));
-            } else {
-                assert!(matches!(vector.get(i), Value::Float64(v) if v == 6.0));
-            }
-        }
-    }
-}