#!/usr/bin/python
# -*- coding: UTF-8 -*-
"""
@author:andrew
@file:ResultSummary.py
@email:admin@marques22.com
@email:2021022362@m.scnu.edu.cn
@time:2023/09/15
"""
import numpy as np
import pandas as pd


def AnalyseSegment(df_segment, thresh=0.5, thresh_event_interval=2, thresh_event_length=8,
                   save_path=None, true_sleep_time=0):
    df_segment["thresh_Pred"] = df_segment["pred"].apply(lambda x: 1 if x > thresh else 0)
    thresh_Pred = df_segment["thresh_Pred"].values
    thresh_Pred2 = thresh_Pred.copy()

    # 扩充
    indices = np.where(thresh_Pred == 1)[0]
    for i in range(len(indices) - 1):
        if indices[i + 1] - indices[i] <= thresh_event_interval:
            thresh_Pred2[indices[i]:indices[i + 1]] = 1
    # 事件判断
    # 如果连续的1的长度大于阈值，则判断为事件，记录起止位置
    diffs = np.diff(thresh_Pred2, prepend=0, append=0)
    start_indices = np.where(diffs == 1)[0]
    end_indices = np.where(diffs == -1)[0]
    event_indices = np.where(end_indices - start_indices >= thresh_event_length)[0]
    result = [(start_indices[i], end_indices[i]) for i in event_indices]

    df_event = pd.DataFrame(result, columns=["start", "end"])
    df_event["length"] = df_event["end"] - df_event["start"]
    df_event["start"] = df_event["start"] - 29
    df_event["end"] = df_event["end"] - 29

    if save_path is not None:
        df_event.to_csv(save_path, index=False)

    # 根据AHI评估严重程度
    if true_sleep_time is 0:
        record_length = len(df_segment) / 60 / 60
        AHI = len(result) / record_length
    else:
        record_length = true_sleep_time / 60
        AHI = len(result) / record_length

    if AHI < 5:
        severity = "Healthy"
    elif AHI < 15:
        severity = "Mild"
    elif AHI < 30:
        severity = "Moderate"
    else:
        severity = "Severe"

    SA_HYP_count = len(result)
    print("patient: ", save_path.stem)
    print("Event number: ", SA_HYP_count)
    print("Record length (hours): ", round(record_length, 2))
    print("AHI: ", AHI)
    print("Severity: ", severity)
    print("=====================================")

    return SA_HYP_count, AHI, severity