MQL5-Google-Onedrive/scripts/ci_validate_repo.py

#!/usr/bin/env python3
"""
Lightweight repository sanity checks suitable for GitHub Actions.
This is intentionally NOT a compiler for MQL5 (MetaEditor isn't available on CI).
"""

from __future__ import annotations

import os
import sys
import time
from pathlib import Path


REPO_ROOT = Path(__file__).resolve().parents[1]
MQL5_DIR = REPO_ROOT / "mt5" / "MQL5"


def fail(msg: str) -> None:
    print(f"ERROR: {msg}", file=sys.stderr)
    raise SystemExit(1)


def iter_source_files() -> list[tuple[Path, os.stat_result]]:
    """⚡ Bolt: Efficiently find source files using os.scandir to avoid Path.rglob overhead."""
    if not MQL5_DIR.exists():
        fail(f"Missing directory: {MQL5_DIR}")

    files: list[tuple[Path, os.stat_result]] = []

    def walk_dir(path: str):
        try:
            with os.scandir(path) as it:
                for entry in it:
                    if entry.is_file():
                        if entry.name.lower().endswith((".mq5", ".mqh")):
                            # ⚡ Bolt: DirEntry.stat() is often cached from the directory scan itself,
                            # especially on Windows, avoiding an extra system call per file.
                            files.append((Path(entry.path), entry.stat()))
                    elif entry.is_dir():
                        walk_dir(entry.path)
        except PermissionError:
            pass

    walk_dir(str(MQL5_DIR))

    if not files:
        fail(f"No .mq5/.mqh files found under {MQL5_DIR}")

    return sorted(files, key=lambda x: x[0])


def validate_files(files: list[tuple[Path, os.stat_result]]) -> None:
    """⚡ Bolt: Single-pass validation using pre-fetched metadata for maximum efficiency."""
    for p, stat in files:
        # ⚡ Bolt: Use pre-fetched stat for early size check before reading file content.
        if stat.st_size > 5_000_000:
            fail(f"Unexpectedly large source file (>5MB): {p.relative_to(REPO_ROOT)} ({stat.st_size} bytes)")

        # ⚡ Bolt: Use chunked binary reading (64KB) to detect NUL bytes efficiently.
        # This minimizes memory overhead compared to full file reads.
        with p.open("rb") as f:
            while chunk := f.read(65536):
                if b"\x00" in chunk:
                    fail(f"NUL byte found in {p.relative_to(REPO_ROOT)}")


def main() -> int:
    start_time = time.perf_counter()

    files = iter_source_files()
    validate_files(files)

    duration = time.perf_counter() - start_time

    rel = [str(p.relative_to(REPO_ROOT)) for p, _ in files]
    print(f"OK: found {len(rel)} source files in {duration:.4f}s:")
    for r in rel:
        print(f"- {r}")
    return 0


if __name__ == "__main__":
    raise SystemExit(main())