"""
批量导入脚本配置文件
统一管理所有配置参数
"""
import os
from pathlib import Path

# 获取项目根目录
PROJECT_ROOT = Path(__file__).parent.parent.parent.absolute()

# 数据文件配置
DATA_ROOT = PROJECT_ROOT / "upload_app" / "data"  # parquet文件根目录
DEFAULT_ACCOUNT_ID = 1  # 默认账号ID

# 批次大小配置（根据数据库性能调整）
BATCH_SIZE = {
    "section": 500,     # 断面数据批次大小
    "checkpoint": 500,  # 观测点批次大小
    "settlement": 500,  # 沉降数据批次大小
    "level": 500,       # 水准数据批次大小
    "original": 1000    # 原始数据批次大小（支持更大批次）
}

# 断点续传配置
RESUME_PROGRESS_FILE = PROJECT_ROOT / "data_import_progress.json"
RESUME_ENABLE = False  # 是否开启断点续传

# 数据类型映射（复用现有逻辑）
DATA_TYPE_MAPPING = {
    "section": (
        "断面数据表",
        "section_",
        "sections",
        ["account_id"]
    ),
    "checkpoint": (
        "观测点数据表",
        "point_",
        "checkpoints",
        []
    ),
    "settlement": (
        "沉降数据表",
        "settlement_",
        "settlements",
        []
    ),
    "level": (
        "水准数据表",
        "level_",
        "levels",
        []
    ),
    "original": (
        "原始数据表",
        "original_",
        "originals",
        []
    )
}

# 数据依赖顺序（可忽略）
DATA_TYPE_ORDER = [
    # ("section", "断面数据"),
    # ("checkpoint", "观测点数据"),
    # ("settlement", "沉降数据"),
    # ("level", "水准数据"),
    ("original", "原始数据")
]

# 核心字段校验配置
CRITICAL_FIELDS = {
    "section": ["section_id", "account_id", "mileage", "work_site"],
    "checkpoint": ["point_id", "section_id", "aname", "burial_date"],
    "settlement": ["NYID", "point_id", "sjName"],
    "level": ["NYID", "linecode", "wsphigh", "createDate"],
    "original": ["NYID", "bfpcode", "mtime", "bfpvalue", "sort"]
}

# 数值型字段强制转换配置
TYPE_CONVERSION = {
    "section": {
        "section_id": int,
        "account_id": int
    },
    "checkpoint": {
        "point_id": int
    },
    "settlement": {
        "NYID": str  # 沉降NYID转为字符串
    },
    "original": {
        "sort": int
    }
}

# 日志配置
LOG_LEVEL = "INFO"
LOG_FORMAT = "%(asctime)s - %(name)s - %(levelname)s - %(message)s"

# 文件过滤配置
MIN_FILE_SIZE = 1024  # 最小文件大小（字节），过滤空文件