tools/mpy-triage: Add database layer and GitHub API wrapper

.github/workflows/ports_windows.yml

@@ -25,23 +25,24 @@ jobs:
         platform: [x86, x64]
         configuration: [Debug, Release]
         variant: [dev, standard]
-        visualstudio: ['2017', '2019', '2022']
+        visualstudio: ['2019', '2022']
         include:
-        - visualstudio: '2017'
-          vs_version: '[15, 16)'
-          custom_vs_install: true
         - visualstudio: '2019'
-          vs_version: '[16, 17)'
-          custom_vs_install: true
+          # The v142 toolset (VS 2019 compiler) is pre-installed on
+          # windows-2022 as a component of VS 2022.  Use VS 2022's
+          # MSBuild and select the v142 toolset via PlatformToolset.
+          vs_version: '[17, 18)'
+          platform_toolset: v142
         - visualstudio: '2022'
           vs_version: '[17, 18)'
+          platform_toolset: v143
         # trim down the number of jobs in the matrix
         exclude:
         - variant: standard
           configuration: Debug
         - visualstudio: '2019'
           configuration: Debug
-    runs-on: windows-latest
+    runs-on: windows-2022
     env:
       CI_BUILD_CONFIGURATION: ${{ matrix.configuration }}
     steps:
@@ -51,31 +52,20 @@ jobs:
       uses: actions/setup-python@v6
       with:
         python-version: '3.11'
-    - name: Install Visual Studio ${{ matrix.visualstudio }}
-      if: matrix.custom_vs_install
-      shell: bash
-      # Shell functions in this block are to retry intermittent corrupt
-      # downloads (with a clean download dir) before failing the job outright
-      run: |
-        try () { ($@) || ($@) || ($@) || ($@) }
-        clean_install () ( rm -rf $TEMP/chocolatey; choco install $1 )
-        try clean_install visualstudio${{ matrix.visualstudio }}buildtools
-        try clean_install visualstudio${{ matrix.visualstudio }}-workload-vctools
-        try clean_install windows-sdk-8.1
     - uses: microsoft/setup-msbuild@v2
       with:
         vs-version: ${{ matrix.vs_version }}
     - uses: actions/checkout@v6
     - name: Build mpy-cross.exe
-      run: msbuild mpy-cross\mpy-cross.vcxproj -maxcpucount -property:Configuration=${{ matrix.configuration }} -property:Platform=${{ matrix.platform }}
+      run: msbuild mpy-cross\mpy-cross.vcxproj -maxcpucount -property:Configuration=${{ matrix.configuration }} -property:Platform=${{ matrix.platform }} -property:PlatformToolset=${{ matrix.platform_toolset }}
     - name: Update submodules
       run: git submodule update --init lib/micropython-lib
     - name: Build micropython.exe
-      run: msbuild ports\windows\micropython.vcxproj -maxcpucount -property:Configuration=${{ matrix.configuration }} -property:Platform=${{ matrix.platform }} -property:PyVariant=${{ matrix.variant }}
+      run: msbuild ports\windows\micropython.vcxproj -maxcpucount -property:Configuration=${{ matrix.configuration }} -property:Platform=${{ matrix.platform }} -property:PyVariant=${{ matrix.variant }} -property:PlatformToolset=${{ matrix.platform_toolset }}
     - name: Get micropython.exe path
       id: get_path
       run: |
-        $exePath="$(msbuild ports\windows\micropython.vcxproj -nologo -v:m -t:ShowTargetPath -property:Configuration=${{ matrix.configuration }} -property:Platform=${{ matrix.platform }} -property:PyVariant=${{ matrix.variant }})"
+        $exePath="$(msbuild ports\windows\micropython.vcxproj -nologo -v:m -t:ShowTargetPath -property:Configuration=${{ matrix.configuration }} -property:Platform=${{ matrix.platform }} -property:PyVariant=${{ matrix.variant }} -property:PlatformToolset=${{ matrix.platform_toolset }})"
         echo ("micropython=" + $exePath.Trim()) >> $env:GITHUB_OUTPUT
     - name: Run tests
       id: test

tools/mpy-triage/.gitignore

@@ -0,0 +1,6 @@
+*.egg-info/
+__pycache__/
+.venv/
+*.pyc
+data/*.db
+.pytest_cache/

tools/mpy-triage/data/schema.sql

@@ -0,0 +1,31 @@
+-- MicroPython Issue Triage Database Schema
+
+-- Issues and PRs
+CREATE TABLE IF NOT EXISTS items (
+    id INTEGER PRIMARY KEY,
+    number INTEGER NOT NULL,
+    repo TEXT NOT NULL DEFAULT 'micropython/micropython',
+    type TEXT NOT NULL,      -- 'issue' or 'pr'
+    title TEXT,
+    body TEXT,
+    author TEXT,
+    state TEXT,              -- open, closed
+    labels TEXT,             -- JSON array of label names
+    created_at TEXT,
+    closed_at TEXT,
+    updated_at TEXT,
+    UNIQUE(repo, number)
+);
+
+-- Sync state for incremental updates
+CREATE TABLE IF NOT EXISTS sync_state (
+    key TEXT PRIMARY KEY,
+    value TEXT
+);
+
+-- Indexes
+CREATE INDEX IF NOT EXISTS idx_items_repo ON items(repo);
+CREATE INDEX IF NOT EXISTS idx_items_type ON items(type);
+CREATE INDEX IF NOT EXISTS idx_items_state ON items(state);
+CREATE INDEX IF NOT EXISTS idx_items_created ON items(created_at);
+CREATE INDEX IF NOT EXISTS idx_items_repo_number ON items(repo, number);

tools/mpy-triage/pyproject.toml

@@ -0,0 +1,32 @@
+[build-system]
+requires = ["setuptools>=61.0", "wheel"]
+build-backend = "setuptools.build_meta"
+
+[project]
+name = "mpy-triage"
+version = "0.1.0"
+description = "CLI tool for detecting duplicate/related MicroPython issues and PRs"
+requires-python = ">=3.10"
+license = {text = "MIT"}
+
+dependencies = [
+    "sqlite-vec>=0.1.6",
+]
+
+[project.optional-dependencies]
+dev = [
+    "pytest>=7.0.0",
+    "ruff>=0.4.0",
+]
+
+[project.scripts]
+mpy-triage = "mpy_triage.cli:main"
+
+[tool.setuptools.packages.find]
+where = ["src"]
+
+[tool.ruff]
+line-length = 99
+
+[tool.pytest.ini_options]
+testpaths = ["tests"]

tools/mpy-triage/src/mpy_triage/cli.py

@@ -0,0 +1,29 @@
+"""CLI entry point for mpy-triage."""
+
+import argparse
+import logging
+
+
+def main() -> None:
+    """Main entry point for the mpy-triage CLI."""
+    parser = argparse.ArgumentParser(
+        prog="mpy-triage",
+        description="Detect duplicate and related MicroPython issues/PRs using semantic search.",
+    )
+    parser.add_argument("-v", "--verbose", action="store_true", help="Enable verbose logging.")
+    parser.add_argument(
+        "--db",
+        default="data/triage.db",
+        help="Path to the SQLite database (default: data/triage.db).",
+    )
+
+    _args = parser.parse_args()
+
+    logging.basicConfig(
+        level=logging.DEBUG if _args.verbose else logging.INFO,
+        format="%(levelname)s: %(message)s",
+    )
+
+
+if __name__ == "__main__":
+    main()

tools/mpy-triage/src/mpy_triage/db.py

@@ -0,0 +1,85 @@
+"""Database layer for mpy-triage.
+
+Manages SQLite connections, schema initialization, sync state, and
+sqlite-vec extension loading.
+"""
+
+import logging
+import sqlite3
+from pathlib import Path
+
+log = logging.getLogger(__name__)
+
+
+def get_connection(db_path: Path) -> sqlite3.Connection:
+    """Open an SQLite connection with Row factory and WAL mode.
+
+    Args:
+        db_path: Path to the SQLite database file.
+
+    Returns:
+        An open sqlite3.Connection.
+    """
+    conn = sqlite3.connect(str(db_path))
+    conn.row_factory = sqlite3.Row
+    conn.execute("PRAGMA journal_mode=WAL")
+    return conn
+
+
+def init_db(conn: sqlite3.Connection, schema_path: Path) -> None:
+    """Read and execute a schema SQL file against the connection.
+
+    Safe to call multiple times (uses IF NOT EXISTS in schema).
+
+    Args:
+        conn: An open sqlite3 connection.
+        schema_path: Path to the .sql schema file.
+    """
+    schema_path = Path(schema_path)
+    schema_sql = schema_path.read_text()
+    conn.executescript(schema_sql)
+    conn.commit()
+
+
+def get_sync_state(conn: sqlite3.Connection, key: str) -> str | None:
+    """Retrieve a value from the sync_state table.
+
+    Args:
+        conn: An open sqlite3 connection.
+        key: The sync state key to look up.
+
+    Returns:
+        The stored value string, or None if the key is not found.
+    """
+    cursor = conn.execute("SELECT value FROM sync_state WHERE key = ?", (key,))
+    row = cursor.fetchone()
+    return row[0] if row else None
+
+
+def set_sync_state(conn: sqlite3.Connection, key: str, value: str) -> None:
+    """Insert or update a value in the sync_state table.
+
+    Args:
+        conn: An open sqlite3 connection.
+        key: The sync state key.
+        value: The value to store.
+    """
+    conn.execute(
+        "INSERT OR REPLACE INTO sync_state (key, value) VALUES (?, ?)",
+        (key, value),
+    )
+    conn.commit()
+
+
+def load_vec_extension(conn: sqlite3.Connection) -> None:
+    """Load the sqlite-vec extension into the connection.
+
+    Args:
+        conn: An open sqlite3 connection.
+    """
+    import sqlite_vec
+
+    conn.enable_load_extension(True)
+    sqlite_vec.load(conn)
+    conn.enable_load_extension(False)
+    log.info("sqlite-vec extension loaded")