import re
import pandas as pd
from pathlib import Path
from typing import List, Dict


def parse_line(line: str) -> Dict:
    """
    解析一行数据，返回一个字典
    例如输入:
    [300]心率:78,83,83,83,;标准差1:43.0491;标准差2:2.16506;均方根:82.2787;值:82;信任级别:1     血氧组:;标准差1:0;标准差:0;均方根:0;值:0;信任级别:0结果延迟:1     R组:;均方根:0;     仪器:60
[325]心率:75,78,83,83,;标准差1:38.6641;标准差2:3.4187;均方根:80.3232;值:81;信任级别:1     血氧组:;标准差1:0;标准差:0;均方根:0;值:0;信任级别:0结果延迟:2     R组:;均方根:0;     仪器:60
[350]心率:78,83,83,83,;标准差1:46.0893;标准差2:2.16506;均方根:82.2787;值:82;信任级别:1     血氧组:;标准差1:0;标准差:0;均方根:0;值:0;信任级别:0结果延迟:3     R组:;均方根:0;     仪器:60
[375]心率:83,83,83,107,;标准差1:43.8422;标准差2:10.3923;均方根:90.1047;值:82;信任级别:0     血氧组:;标准差1:0;标准差:0;均方根:0;值:0;信任级别:0结果延迟:3     R组:;均方根:0;     仪器:60
[400]心率:83,83,107,107,;标准差1:47.7501;标准差2:12;均方根:96.2549;值:82;信任级别:0     血氧组:;标准差1:0;标准差:0;均方根:0;值:0;信任级别:0结果延迟:3     R组:;均方根:0;     仪器:60
    """
    line = line.strip()
    if not line:
        return None

    # 提取时间戳 [300] 这样的
    time_match = re.match(r'\[(\d+)\]', line)
    if not time_match:
        return None
    timestamp = int(time_match.group(1))

    # 去掉时间戳部分，后面全部是键值对
    content = line[time_match.end():]
    content = content.split("     ")

    result = {"timestamp": timestamp}

    for block in content:
        parsed_block = parse_block(block)
        result.update(parsed_block)
    return result


def parse_block(block: str) -> Dict:
    """
    解析一个数据块，返回一个字典
    例如输入:
    心率:78,83,83,83,;标准差1:43.0491;标准差2:2.16506;均方根:82.2787;值:82;信任级别:1
    血氧组:;标准差1:0;标准差:0;均方根:0;值:0;信任级别:0结果延迟:1
    R组:;均方根:0;
    仪器:60
    """
    data = {}
    parts = block.split(';')
    if len(parts) == 1:
        key, value = parts[0].split(':', 1)
        data[key.strip()] = value.strip()
    else:
        key_0 = parts[0].split(':', 1)[0].strip()
        for part in parts:
            if len(part.strip()) == 0:
                continue
            if "结果延迟" in part:
                part0, key1, value1 = part.partition('结果延迟:')
                data[f"{key_0}_结果延迟"] = value1.strip()
                key2, _, value2 = part0.partition(':')
                if key2 and value2:
                    data[f"{key_0}_{key2.strip()}"] = value2.strip()
            else:
                key, value = part.split(':', 1)
                data[f"{key_0}_{key.strip()}"] = value.strip()

    return data



def parse_dat_file(file_path) -> pd.DataFrame:
    """
    读取整个 .dat 文件并返回 pandas DataFrame
    """
    data = []
    with open(file_path, 'r', encoding='utf-8') as f:
        for line_num, line in enumerate(f, 1):
            parsed = parse_line(line)
            if parsed:
                data.append(parsed)
            else:
                if line.strip():
                    print(f"第 {line_num} 行解析失败，已跳过: {line.strip()[:80]}...")

    df = pd.DataFrame(data)

    # 按时间戳排序（一般已经是顺序，但保险起见）
    if not df.empty:
        df = df.sort_values("timestamp").reset_index(drop=True)

    return df

if __name__ == '__main__':
    # file_path = "./data/spo2_alg_info[42-A3-C5-2F-F7-32]2025.11.17_17.01.14.dat"
    file_path = "./data/spo2_alg_info[42_A3_C5_2F_F7_32]2025.11.17_17.12.16.dat"
    file_path = Path(file_path)
    df = parse_dat_file(file_path)

    df.to_csv(file_path.with_suffix('.csv'), index=False, encoding='utf-8-sig')