MeshSate/app/diagnostic/config.py

"""MeshSafe Diagnostics Kit — 配置读取。

所有配置项可被环境变量覆盖（``MESHSAFE_DIAG_*`` 前缀），
优先级：环境变量 > diagnostics.yaml > 默认值。
"""

from __future__ import annotations

import functools
import os
from pathlib import Path
from typing import Any, Dict

import yaml


_PROJECT_ROOT = Path(__file__).resolve().parent.parent.parent
_DEFAULT_YAML = _PROJECT_ROOT / "config" / "diagnostics.yaml"


def _load_yaml(path: Path) -> Dict[str, Any]:
    """读取 YAML 配置，文件不存在返回空 dict。"""
    try:
        with open(path, "r", encoding="utf-8") as f:
            data = yaml.safe_load(f) or {}
        return data.get("diagnostics", {}) if isinstance(data, dict) else {}
    except FileNotFoundError:
        return {}
    except Exception:  # noqa: BLE001
        return {}


def _env_bool(name: str, default: bool) -> bool:
    v = os.environ.get(name)
    if v is None:
        return default
    return v.strip().lower() in ("1", "true", "yes", "on")


def _env_int(name: str, default: int) -> int:
    v = os.environ.get(name)
    if v is None:
        return default
    try:
        return int(v)
    except ValueError:
        return default


def _env_str(name: str, default: str) -> str:
    return os.environ.get(name, default)


@functools.lru_cache(maxsize=1)
def get_diag_config() -> "DiagConfig":
    yaml_data = _load_yaml(_DEFAULT_YAML)
    return DiagConfig.from_dict(yaml_data)


class DiagConfig:
    """诊断套件配置。"""

    def __init__(self, yaml_data: Dict[str, Any]):
        # 顶层开关
        yaml_enabled = yaml_data.get("enabled", True)
        self.enabled = _env_bool(
            "MESHSAFE_DIAG_ENABLED", yaml_enabled
        )

        # Docker 子能力
        docker_yaml = yaml_data.get("docker", {})
        yaml_docker_mode = docker_yaml.get("enabled", "auto")
        self.docker_mode = _env_str(
            "MESHSAFE_DIAG_DOCKER", yaml_docker_mode
        ).lower()
        if self.docker_mode not in ("auto", "true", "false"):
            self.docker_mode = "auto"

        # 定时器周期（秒）
        timers_yaml = yaml_data.get("timers", {})
        self.error_scan_interval = _env_int(
            "MESHSAFE_DIAG_ERROR_INTERVAL",
            timers_yaml.get("error_scan_interval", 30),
        )
        self.container_check_interval = _env_int(
            "MESHSAFE_DIAG_CONTAINER_INTERVAL",
            timers_yaml.get("container_check_interval", 300),
        )
        self.db_aggregate_interval = _env_int(
            "MESHSAFE_DIAG_DB_INTERVAL",
            timers_yaml.get("db_aggregate_interval", 60),
        )

        # 日志行数
        self.docker_log_tail = int(yaml_data.get("docker_log_tail", 500))
        self.export_log_tail = int(yaml_data.get("export_log_tail", 2000))

        # 导出包数据范围
        export_yaml = yaml_data.get("export", {})
        self.export_db_span_days = int(export_yaml.get("db_span_days", 7))
        self.export_scan_tasks_max = int(export_yaml.get("scan_tasks_max", 10000))
        self.export_scan_findings_max = int(export_yaml.get("scan_findings_max", 50000))
        self.export_engine_errors_max = int(export_yaml.get("engine_errors_max", 5000))

        # 规则阈值
        rules_yaml = yaml_data.get("rules", {})
        self.error_ratio_threshold = float(
            rules_yaml.get("error_ratio_threshold", 0.05)
        )
        self.clamav_db_stale_days = int(
            rules_yaml.get("clamav_db_stale_days", 7)
        )
        self.reject_ratio_threshold = float(
            rules_yaml.get("reject_ratio_threshold", 0.20)
        )
        self.suspicious_exceeds_malicious = bool(
            rules_yaml.get("suspicious_exceeds_malicious", True)
        )

        # 持久化存储
        storage_yaml = yaml_data.get("storage", {})
        self.storage_enabled = _env_bool(
            "MESHSAFE_DIAG_STORAGE_ENABLED",
            storage_yaml.get("enabled", True),
        )
        self.storage_dir = storage_yaml.get("dir", "/app/data/diagnostics")
        self.history_retention_days = int(
            storage_yaml.get("history_retention_days", 7)
        )
        self.log_snapshot_on_timer = bool(
            storage_yaml.get("log_snapshot_on_timer", True)
        )
        self.log_snapshot_tail = int(
            storage_yaml.get("log_snapshot_tail", 500)
        )

    @classmethod
    def from_dict(cls, data: Dict[str, Any]) -> "DiagConfig":
        return cls(data)

    def to_dict(self) -> Dict[str, Any]:
        return {
            "enabled": self.enabled,
            "docker_mode": self.docker_mode,
            "timers": {
                "error_scan_interval": self.error_scan_interval,
                "container_check_interval": self.container_check_interval,
                "db_aggregate_interval": self.db_aggregate_interval,
            },
            "docker_log_tail": self.docker_log_tail,
            "export_log_tail": self.export_log_tail,
            "export": {
                "db_span_days": self.export_db_span_days,
                "scan_tasks_max": self.export_scan_tasks_max,
                "scan_findings_max": self.export_scan_findings_max,
                "engine_errors_max": self.export_engine_errors_max,
            },
            "rules": {
                "error_ratio_threshold": self.error_ratio_threshold,
                "clamav_db_stale_days": self.clamav_db_stale_days,
                "reject_ratio_threshold": self.reject_ratio_threshold,
                "suspicious_exceeds_malicious": self.suspicious_exceeds_malicious,
            },
            "storage": {
                "enabled": self.storage_enabled,
                "dir": self.storage_dir,
                "history_retention_days": self.history_retention_days,
                "log_snapshot_on_timer": self.log_snapshot_on_timer,
                "log_snapshot_tail": self.log_snapshot_tail,
            },
        }