import hashlib
import json
import os
import re
import sys
from typing import Dict, List


class SourceRange(dict):
    def __init__(self, begin: int, end: int, column: int):
        super().__init__(begin=begin, end=end, column=column)


class SourceLocation(dict):
    def __init__(self, path: str, lines: Dict[str, int]):
        super().__init__(path=path, lines=lines)


class ClangTidyError(dict):
    def __init__(
        self,
        description: str,
        check_name: str,
        fingerprint: str,
        severity: str,
        location: SourceLocation,
    ):
        super().__init__(
            description=description,
            check_name=check_name,
            fingerprint=fingerprint,
            severity=severity,
            location=location,
        )


def parse_clang_tidy(log_file_path: str) -> List[ClangTidyError]:
    issues = []
    pattern = re.compile(
        r"^(.*?):(\d+):(\d+):\s+(warning|error|note):\s+(.*?)\s+\[(.*?)\]$"
    )
    repo_root = os.environ.get("CI_PROJECT_DIR", os.getcwd())

    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"

            try:
                path = os.path.relpath(path, repo_root)
            except ValueError:
                pass

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

            issues.append(
                ClangTidyError(
                    description=f"{message}",
                    check_name=name,
                    fingerprint=fingerprint,
                    severity=severity,
                    location=SourceLocation(
                        path=path,
                        lines=SourceRange(
                            begin=int(line),
                            end=int(line),
                            column=int(column),
                        ),
                    ),
                )
            )

    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))