chore(release): Querya Desktop 0.4.19

.devcontainer/Dockerfile

@@ -0,0 +1,14 @@
+FROM ghcr.io/cirruslabs/flutter:3.41.6
+
+# Linux desktop toolchain + libs the app links against.
+RUN apt-get update \
+ && apt-get install -y --no-install-recommends \
+      clang cmake ninja-build pkg-config libgtk-3-dev libsecret-1-dev \
+      libsqlite3-dev libglu1-mesa xvfb git curl ca-certificates \
+ && rm -rf /var/lib/apt/lists/*
+
+# Docker CLI so the dev stack (docker/docker-compose.yml) can be started from inside the container.
+COPY --from=docker:27-cli /usr/local/bin/docker /usr/local/bin/docker
+COPY --from=docker:27-cli /usr/local/libexec/docker/cli-plugins/docker-compose /usr/local/libexec/docker/cli-plugins/docker-compose
+
+ENV DISPLAY=:99

.devcontainer/devcontainer.json

@@ -0,0 +1,23 @@
+{
+  "name": "Querya Desktop",
+  "build": { "dockerfile": "Dockerfile" },
+  "features": {
+    "ghcr.io/devcontainers/features/docker-outside-of-docker:1": {}
+  },
+  "postCreateCommand": "flutter config --enable-linux-desktop && flutter pub get",
+  "postStartCommand": "docker compose -f docker/docker-compose.yml up -d || echo 'Docker stack not started (run it manually: cd docker && docker compose up -d)'",
+  "forwardPorts": [5432, 3306, 6379, 27017, 8123, 9000],
+  "portsAttributes": {
+    "5432": { "label": "PostgreSQL" },
+    "3306": { "label": "MySQL" },
+    "6379": { "label": "Redis" },
+    "27017": { "label": "MongoDB" },
+    "8123": { "label": "ClickHouse HTTP" }
+  },
+  "customizations": {
+    "vscode": {
+      "extensions": ["Dart-Code.dart-code", "Dart-Code.flutter"]
+    }
+  },
+  "hostRequirements": { "cpus": 4, "memory": "8gb" }
+}

.github/actions/test-extension/action.yml

@@ -0,0 +1,128 @@
+name: Test Querya extension
+description: >-
+  Downloads a released querya-ext-tester and checks a Querya database-driver
+  extension (manifest and JSON-RPC lifecycle) headlessly, with no Flutter
+  install. Publish a copy of this directory as QueryaHub/action-test-extension
+  to use it as `uses: QueryaHub/action-test-extension@v1`.
+
+inputs:
+  querya-version:
+    description: Querya Desktop release to take the tester from (e.g. 0.4.18) or "latest".
+    required: false
+    default: latest
+  target-version:
+    description: Host version to emulate (--target-version). Defaults to querya-version.
+    required: false
+    default: ''
+  extension-dir:
+    description: Directory containing manifest.json and the driver entry point.
+    required: false
+    default: .
+  connect:
+    description: db.connect parameters as inline JSON or a path to a JSON file.
+    required: false
+    default: ''
+  query:
+    description: Statement for the db.query check.
+    required: false
+    default: SELECT 1
+  table:
+    description: Table name for the db.getTableSchema check (skipped when empty).
+    required: false
+    default: ''
+  sandbox:
+    description: Run the plugin inside the OS sandbox when available ("true" / "false").
+    required: false
+    default: 'false'
+  junit-path:
+    description: Where to write the JUnit XML report.
+    required: false
+    default: querya-ext-test-junit.xml
+  tap-path:
+    description: Where to write the TAP report (skipped when empty).
+    required: false
+    default: ''
+
+outputs:
+  tester-version:
+    description: Querya Desktop release the tester binary came from.
+    value: ${{ steps.fetch.outputs.version }}
+
+runs:
+  using: composite
+  steps:
+    - name: Download querya-ext-tester
+      id: fetch
+      shell: bash
+      env:
+        GH_TOKEN: ${{ github.token }}
+        REQUESTED: ${{ inputs.querya-version }}
+        REPO: QueryaHub/Querya-Desktop
+      run: |
+        set -euo pipefail
+        case "${RUNNER_OS}-${RUNNER_ARCH}" in
+          Linux-X64)    TARGET=linux-x64;   EXT=tar.gz ;;
+          macOS-ARM64)  TARGET=macos-arm64; EXT=tar.gz ;;
+          macOS-X64)    TARGET=macos-x64;   EXT=tar.gz ;;
+          Windows-X64)  TARGET=windows-x64; EXT=zip ;;
+          *) echo "::error::Unsupported runner ${RUNNER_OS}-${RUNNER_ARCH}"; exit 1 ;;
+        esac
+
+        if [ "$REQUESTED" = "latest" ]; then
+          TAG=$(gh api "repos/${REPO}/releases/latest" --jq .tag_name)
+        else
+          TAG="$REQUESTED"
+        fi
+        VERSION="${TAG#v}"
+        ASSET="querya-ext-tester-v${VERSION}-${TARGET}.${EXT}"
+
+        DEST="${RUNNER_TEMP}/querya-ext-tester"
+        rm -rf "$DEST" && mkdir -p "$DEST"
+        gh release download "$TAG" --repo "$REPO" --dir "$DEST" \
+          --pattern "$ASSET" --pattern SHA256SUMS.txt
+
+        cd "$DEST"
+        EXPECTED=$(grep " ${ASSET}\$" SHA256SUMS.txt | cut -d' ' -f1)
+        if [ -z "$EXPECTED" ]; then
+          echo "::error::${ASSET} is not listed in SHA256SUMS.txt of release ${TAG}"
+          exit 1
+        fi
+        if command -v sha256sum >/dev/null; then
+          ACTUAL=$(sha256sum "$ASSET" | cut -d' ' -f1)
+        else
+          ACTUAL=$(shasum -a 256 "$ASSET" | cut -d' ' -f1)
+        fi
+        if [ "$EXPECTED" != "$ACTUAL" ]; then
+          echo "::error::Checksum mismatch for ${ASSET}"
+          exit 1
+        fi
+
+        if [ "$EXT" = "zip" ]; then 7z x -y "$ASSET" >/dev/null; else tar -xzf "$ASSET"; fi
+        echo "dir=${DEST}" >> "$GITHUB_OUTPUT"
+        echo "version=${VERSION}" >> "$GITHUB_OUTPUT"
+
+    - name: Run querya-ext-tester
+      shell: bash
+      env:
+        TESTER_DIR: ${{ steps.fetch.outputs.dir }}
+        TARGET_VERSION: ${{ inputs.target-version != '' && inputs.target-version || steps.fetch.outputs.version }}
+        EXTENSION_DIR: ${{ inputs.extension-dir }}
+        CONNECT: ${{ inputs.connect }}
+        QUERY: ${{ inputs.query }}
+        TABLE: ${{ inputs.table }}
+        SANDBOX: ${{ inputs.sandbox }}
+        JUNIT_PATH: ${{ inputs.junit-path }}
+        TAP_PATH: ${{ inputs.tap-path }}
+      run: |
+        set -euo pipefail
+        EXE="${TESTER_DIR}/querya-ext-tester"
+        if [ "${RUNNER_OS}" = "Windows" ]; then EXE="${EXE}.exe"; fi
+
+        ARGS=("$EXTENSION_DIR" --target-version "$TARGET_VERSION" --query "$QUERY" --no-color)
+        if [ -n "$CONNECT" ]; then ARGS+=(--connect "$CONNECT"); fi
+        if [ -n "$TABLE" ]; then ARGS+=(--table "$TABLE"); fi
+        if [ "$SANDBOX" = "true" ]; then ARGS+=(--sandbox); fi
+        if [ -n "$JUNIT_PATH" ]; then ARGS+=(--junit "$JUNIT_PATH"); fi
+        if [ -n "$TAP_PATH" ]; then ARGS+=(--tap "$TAP_PATH"); fi
+
+        "$EXE" "${ARGS[@]}"

.github/workflows/benchmark.yml

@@ -0,0 +1,139 @@
+name: Benchmark
+
+# Informational: compares frame timings of benchmark/grid_perf_bench.dart on the
+# PR against its base branch, three interleaved runs per side, and comments when
+# the Dart build time is slower in every pair (scripts/ci/bench_compare.py).
+# Never blocks a merge.
+
+on:
+  pull_request:
+    branches: [ dev, main ]
+    paths:
+      - 'lib/features/workspace/**'
+      - 'lib/core/database/**'
+      - 'lib/core/csv/**'
+      - 'lib/features/erd/**'
+      - 'lib/features/results/**'
+      - 'benchmark/**'
+      - 'scripts/ci/bench_compare.py'
+      - 'scripts/ci/app_bench_report.py'
+      - '.github/workflows/benchmark.yml'
+  workflow_dispatch:
+
+permissions:
+  contents: read
+  pull-requests: write
+
+concurrency:
+  group: benchmark-${{ github.ref }}
+  cancel-in-progress: true
+
+jobs:
+  grid-bench:
+    name: Grid performance
+    # Forks get no write token and must never reach any self-hosted runner.
+    if: github.event_name != 'pull_request' || github.event.pull_request.head.repo.full_name == github.repository
+    runs-on: ubuntu-latest
+    timeout-minutes: 75
+    continue-on-error: true
+    steps:
+      - uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+
+      - name: Install Linux dependencies
+        timeout-minutes: 8
+        run: |
+          sudo apt-get -o Acquire::Retries=3 update
+          sudo apt-get -o Acquire::Retries=3 install -y clang cmake ninja-build pkg-config libgtk-3-dev libsecret-1-dev libsqlite3-dev xvfb libgl1-mesa-dri
+
+      - uses: subosito/flutter-action@v2
+        with:
+          flutter-version: '3.47.1'
+          channel: 'stable'
+          cache: true
+
+      - name: Prepare the PR and the base checkouts
+        run: |
+          flutter pub get
+          if [ "${{ github.event_name }}" = "pull_request" ]; then
+            git worktree add "$RUNNER_TEMP/base" "origin/${{ github.base_ref }}"
+            # The base runs the PR's benchmark, so both sides measure the same thing.
+            rm -rf "$RUNNER_TEMP/base/benchmark"
+            cp -r benchmark "$RUNNER_TEMP/base/benchmark"
+            (cd "$RUNNER_TEMP/base" && flutter pub get)
+          fi
+
+      # Base and PR alternate (base, PR, base, PR, base, PR), so a slow
+      # stretch of the shared runner lands on both sides.
+      - name: Run the benchmark, 3 interleaved runs per side
+        timeout-minutes: 40
+        run: |
+          bench() {
+            (cd "$1" && xvfb-run -a -s '-screen 0 1920x1080x24' flutter run --profile -d linux \
+              -t benchmark/grid_perf_bench.dart --dart-define=MODE=scroll \
+              2>&1 | tee "$2") || true
+          }
+          for i in 1 2 3; do
+            if [ "${{ github.event_name }}" = "pull_request" ]; then
+              bench "$RUNNER_TEMP/base" "$RUNNER_TEMP/base$i.log"
+            fi
+            bench "$GITHUB_WORKSPACE" "$RUNNER_TEMP/head$i.log"
+          done
+
+      - name: Compare
+        id: compare
+        run: |
+          python3 scripts/ci/bench_compare.py \
+            --base "$RUNNER_TEMP"/base*.log --head "$RUNNER_TEMP"/head*.log \
+            > "$RUNNER_TEMP/report.txt"
+          cat "$RUNNER_TEMP/report.txt"
+          echo "regression=$(head -1 "$RUNNER_TEMP/report.txt" | cut -d= -f2)" >> "$GITHUB_OUTPUT"
+          tail -n +2 "$RUNNER_TEMP/report.txt" > "$RUNNER_TEMP/report.md"
+          cat "$RUNNER_TEMP/report.md" >> "$GITHUB_STEP_SUMMARY"
+
+      - name: Comment on regression
+        if: github.event_name == 'pull_request' && steps.compare.outputs.regression == '1'
+        env:
+          GH_TOKEN: ${{ github.token }}
+        run: gh pr comment "${{ github.event.pull_request.number }}" --body-file "$RUNNER_TEMP/report.md"
+
+  # #1173: ERD, charts / groupings and multi-tab SQL scenarios of the real app
+  # (benchmark/app_perf_bench.dart, Demo Playground, no database server),
+  # checked against per-scenario limits. Informational, like the grid bench.
+  app-bench:
+    name: App scenarios performance
+    if: github.event_name != 'pull_request' || github.event.pull_request.head.repo.full_name == github.repository
+    runs-on: ubuntu-latest
+    timeout-minutes: 30
+    continue-on-error: true
+    steps:
+      - uses: actions/checkout@v4
+
+      - name: Install Linux dependencies
+        timeout-minutes: 8
+        run: |
+          sudo apt-get -o Acquire::Retries=3 update
+          sudo apt-get -o Acquire::Retries=3 install -y clang cmake ninja-build pkg-config libgtk-3-dev libsecret-1-dev libsqlite3-dev xvfb libgl1-mesa-dri
+
+      - uses: subosito/flutter-action@v2
+        with:
+          flutter-version: '3.47.1'
+          channel: 'stable'
+          cache: true
+
+      - name: Run the app scenarios
+        timeout-minutes: 20
+        run: |
+          flutter pub get
+          xvfb-run -a -s '-screen 0 1920x1080x24' flutter run --profile -d linux \
+            -t benchmark/app_perf_bench.dart \
+            --dart-define=ONLY=charts_5000,groupings_5000,multi_tab_run,erd_open_200,erd_hover,erd_drag,erd_auto_layout \
+            --dart-define=STREAMS= \
+            2>&1 | tee "$RUNNER_TEMP/app.log" || true
+
+      - name: Report
+        run: |
+          python3 scripts/ci/app_bench_report.py "$RUNNER_TEMP/app.log" > "$RUNNER_TEMP/app_report.txt"
+          cat "$RUNNER_TEMP/app_report.txt"
+          tail -n +2 "$RUNNER_TEMP/app_report.txt" >> "$GITHUB_STEP_SUMMARY"