import hashlib
import json
import re
import sys


def parse_clang_tidy(log_file_path: str) -> list[dict]:
    issues = []
    pattern = re.compile(
        r"^(.*?):(\d+):(\d+):\s+(warning|error|note):\s+(.*?)\s+\[(.*?)\]$"
    )

    with open(log_file_path, "r") as f:
        for line in f:
            match = pattern.match(line)
            if not match:
                continue

            path, line, column, severity, message, name = match.groups()
            severity = "minor" if severity == "warning" else "major"

            fingerprint_data = f"{path}:{line}:{name}"
            fingerprint = hashlib.sha256(fingerprint_data.encode()).hexdigest()

            issues.append(
                {
                    "description": f"{message} ({name})",
                    "fingerprint": fingerprint,
                    "severity": severity,
                    "location": {
                        "path": path,
                        "lines": {
                            "begin": int(line),
                        },
                    },
                }
            )

    return issues


if __name__ == "__main__":
    if len(sys.argv) < 2:
        sys.exit("Usage: python3 parse_clang_tidy.py <CLANG_TIDY_LOG_PATH>")

    issues = parse_clang_tidy(sys.argv[1])
    print(json.dumps(issues, indent=2))