# coding: utf-8
from os import scandir
from os.path import isfile, exists
from sys import argv
from openpyxl import load_workbook
from time import time, sleep, strftime, localtime
from threading import Thread
from pandas import read_csv


class GetThreadResult(Thread):
    def __init__(self, func, args=()):
        super(GetThreadResult, self).__init__()
        self.func = func
        self.args = args
        self.result = 0

    def run(self):
        sleep(1)
        self.result = self.func(*self.args)

    def get_result(self):
        Thread.join(self)  # 等待线程执行完毕
        try:
            return self.result
        except Exception as Err:
            return None


def data_process(result_file, raw_data_dirs, av, rr, axis, vel, trq, w2t, rpm):
    # 功能：完成一个结果文件的数据处理
    # 参数：结果文件，数据目录，以及预读取的参数
    # 返回值：-
    file_name = result_file.split('\\')[-1]
    w2t(f"打开文件 {file_name} 需要 1min 左右", 1)

    global stop
    stop = 0
    t_excel = GetThreadResult(load_workbook, args=(result_file, ))
    t_wait = Thread(target=w2t_local, args=('.', 1, w2t))
    t_excel.start()
    t_wait.start()
    t_excel.join()
    wb_result = t_excel.get_result()
    stop = 1
    sleep(1.1)
    w2t('')

    prefix = result_file.split('\\')[-1].split('_')[0]
    for raw_data_dir in raw_data_dirs:
        if raw_data_dir.split('\\')[-1].split('_')[0] == prefix:
            now_doing_msg(raw_data_dir, 'start', w2t)
            _, data_files = traversal_files(raw_data_dir, w2t)
            # 数据文件串行处理模式---------------------------------
            # count = 1
            # for data_file in data_files:
            #     now_doing_msg(data_file, 'start', w2t)
            #     single_file_process(data_file, wb_result, count, av, rr, axis, vel, trq, w2t, rpm)
            #     count += 1
            #     now_doing_msg(data_file, 'done', w2t)
            # ---------------------------------------------------
            # 数据文件并行处理模式---------------------------------
            threads = [Thread(target=single_file_process, args=(data_files[0], wb_result, 1, av, rr, axis, vel, trq, w2t, rpm)),
                       Thread(target=single_file_process, args=(data_files[1], wb_result, 2, av, rr, axis, vel, trq, w2t, rpm)),
                       Thread(target=single_file_process, args=(data_files[2], wb_result, 3, av, rr, axis, vel, trq, w2t, rpm))]
            [t.start() for t in threads]
            [t.join() for t in threads]
            # ---------------------------------------------------
            now_doing_msg(raw_data_dir, 'done', w2t)

    now_doing_msg(result_file, 'done', w2t)
    w2t(f"保存文件 {file_name} 需要 1min 左右", 1)
    stop = 0
    t_excel = Thread(target=wb_result.save, args=(result_file, ))
    t_wait = Thread(target=w2t_local, args=('.', 1, w2t))
    t_excel.start()
    t_wait.start()
    t_excel.join()
    stop = 1
    sleep(1.1)
    w2t('\n')


def check_files(raw_data_dirs, result_files, w2t):
    # 功能：检查数据文件以及结果文件的合规性
    # 参数：数据文件夹，结果文件
    # 返回值：-
    if len(result_files) != 3:
        msg = "结果文件数目错误，结果文件有且只有三个，请确认！"
        for result_file in result_files:
            w2t(result_file)
        w2t(msg)

    prefix = []
    for result_file in result_files:
        prefix.append(result_file.split('\\')[-1].split('_')[0])
    if not sorted(prefix) == sorted(['load33', 'load66', 'load100']):
        wd = result_files[0].split('\\')
        del wd[-1]
        wd = '\\'.join(wd)
        msg = f"""请关闭所有相关数据文件，并检查工作目录 {wd} 下，有且只允许有类似如下三个文件：
              1. load33_自研_制动性能测试.xlsx
              2. load66_自研_制动性能测试.xlsx
              3. load100_自研_制动性能测试.xlsx"""
        w2t(msg)

    for raw_data_dir in raw_data_dirs:
        components = raw_data_dir.split('\\')[-1].split('_')
        sorted(components)
        if components[0] not in ['load33', 'load66', 'load100'] or \
                components[1] not in ['speed33', 'speed66', 'speed100'] or \
                components[2] not in ['reach33', 'reach66', 'reach100']:
            msg = f"报错信息：数据目录 {raw_data_dir} 命名不合规，请参考如下形式\n" \
                  f"命名规则：\n    1. loadAA_speedBB_reachCC\n    2. loadAA_reachBB_speedCC\n" \
                  f"规则解释：AA/BB/CC 指的是负载/速度/臂展的比例\n" \
                  f"load66_speed100_reach33：66% 负载，100% 速度以及 33% 臂展情况下的测试结果文件夹"
            w2t(msg)

        _, raw_data_files = traversal_files(raw_data_dir, w2t)
        if len(raw_data_files) != 3:
            msg = f"数据目录 {raw_data_dir} 下数据文件个数错误，每个数据目录下有且只能有三个以 .data 为后缀的数据文件"
            w2t(msg)
        for raw_data_file in raw_data_files:
            if not (raw_data_file.split('\\')[-1].endswith('.data') or raw_data_file.split('\\')[-1].endswith('.csv')):
                msg = f"数据文件 {raw_data_file} 后缀错误，每个数据目录下有且只能有三个以 .data/csv 为后缀的数据文件"
                w2t(msg)

    w2t("数据目录合规性检查结束，未发现问题......")


def now_doing_msg(docs, flag, w2t):
    # 功能：输出正在处理的文件或目录
    # 参数：文件或目录，start 或 done 标识
    # 返回值：-
    now = strftime('%Y-%m-%d %H:%M:%S', localtime(time()))
    file_type = 'file' if isfile(docs) else 'dir'
    if flag == 'start' and file_type == 'dir':
        w2t(f"[{now}] 正在处理目录 {docs} 中的数据......")
    elif flag == 'start' and file_type == 'file':
        w2t(f"[{now}] 正在处理文件 {docs} 中的数据......")
    elif flag == 'done' and file_type == 'dir':
        w2t(f"[{now}] 目录 {docs} 数据文件已处理完毕......")
    elif flag == 'done' and file_type == 'file':
        w2t(f"[{now}] 文件 {docs} 数据已处理完毕......")


def w2t_local(msg, wait, w2t):
    while True:
        global stop
        if stop == 0 and wait != 0:
            sleep(1)
            w2t(msg, wait)
        else:
            break


def single_file_process(data_file, wb_result, count, av, rr, axis, vel, trq, w2t, rpm):
    # 功能：完成单个数据文件的处理
    # 参数：如上
    # 返回值：-
    if data_file.endswith('.data'):
        sep = '\t'
        df = read_csv(data_file, sep=sep)
    elif data_file.endswith('.csv'):
        sep = ','
        df = read_csv(data_file, sep=sep, encoding='gbk', header=8)

    conditions = sorted(data_file.split('\\')[-2].split('_')[1:])
    result_sheet_name = find_result_sheet_name(conditions, count)
    ws_result = wb_result[result_sheet_name]

    row_max, row_start = find_row_start(data_file, df, conditions, av, rr, axis, vel, w2t, rpm)

    copy_data_to_result(df, ws_result, row_max, row_start, vel, trq, rpm)
    ws_result["C2"] = int(2)
    ws_result["G2"] = int(10+4)


def copy_data_to_result(df, ws_result, row_max, row_start, vel, trq, rpm):
    # 功能：将数据文件中有效数据拷贝至结果文件对应的 sheet
    # 参数：如上
    # 返回值：-
    # 结果文件数据清零
    for row in ws_result.iter_rows(min_row=2, min_col=1, max_row=2000, max_col=2):
        for cell in row:
            cell.value = None

    # 将合适的数据复制到结果文件
    row_max = row_start + 399 if row_max-row_start > 400 else row_max
    rc = 1 if rpm == 1 else 1000

    data = []
    for i in range(row_start, row_max+1):
        data.append(df.iloc[i, vel-1] * rpm)
        data.append(df.iloc[i, trq-1] * rc)

    i = 0
    for row in ws_result.iter_rows(min_row=2, min_col=1, max_row=row_max - row_start + 2, max_col=2):
        for cell in row:
            cell.value = data[i]
            i = i + 1


def find_result_sheet_name(conditions, count):
    # 功能：获取结果文件准确的sheet页名称
    # 参数：臂展和速度的列表
    # 返回值：结果文件对应的sheet name
    # 33%臂展_33%速度_正1
    reach = conditions[0].removeprefix('reach')
    speed = conditions[1].removeprefix('speed')
    result_sheet_name = f"{reach}%臂展_{speed}%速度_正{count}"

    return result_sheet_name


def find_row_start(data_file, df, conditions, av, rr, axis, vel, w2t, rpm):
    # 功能：查找数据文件中有效数据的行号，也即最后一个速度下降的点位
    # 参数：如上
    # 返回值：速度下降点位，最后的数据点位
    ratio = float(conditions[1].removeprefix('speed'))/100
    speed_max = av * ratio * rr / 6
    row_max = row_start = df.index[-1]

    threshold, step = get_threshold_step(data_file, axis)
    while row_start > step+1:
        speed = df.iloc[row_start, vel-1] * rpm
        if int(speed) < 1:
            row_start -= 50
            continue

        _ = []
        for i in range(row_start, row_start-step+1, -1):
            _.append(df.iloc[i, vel-1] * rpm)
        speed_avg = abs(sum(_))/len(_)

        if abs(speed_avg-speed_max) < threshold:
            row_start = row_start - 10
            break
        else:
            row_start -= step
    else:
        msg = f"可能是{data_file}这个文件数据采集有问题，比如未采集理论速度值，也有可能是程序步长设定问题，请检查......"
        w2t(msg)

    return row_max, row_start


def get_threshold_step(excel_file, axis):
    # 功能：负载和速度100%，且是j2的时候，做特殊处理
    # 参数：新生成的excel，轴号
    # 返回值：速度差阈值，处理步长
    conditions = sorted(excel_file.split('\\')[-2].split('_'))
    # 只有负载和速度是100%时，才会启用更敏感的step
    flg = 1 if conditions[0][-3:] == '100' and conditions[2][-3:] == '100' else 0
    if flg == 1 and axis == 2:
        threshold = 30
        step = 5
    else:
        threshold = 10
        step = 5

    return threshold, step


def traversal_files(path, w2t):
    # 功能：以列表的形式分别返回指定路径下的文件和文件夹，不包含子目录
    # 参数：路径
    # 返回值：路径下的文件夹列表 路径下的文件列表
    if not exists(path):
        msg = f'数据文件夹{path}不存在，请确认后重试......'
        w2t(msg)
    else:
        dirs = []
        files = []
        for item in scandir(path):
            if item.is_dir():
                dirs.append(item.path)
            elif item.is_file():
                files.append(item.path)

        return dirs, files


def main(path, av, rr, rpm, axis, vel, trq, w2t):
    # 功能：执行处理所有数据文件
    # 参数：initialization函数的返回值
    # 返回值：-
    time_start = time()

    raw_data_dirs, result_files = traversal_files(path, w2t)
    check_files(raw_data_dirs, result_files, w2t)

    prefix = []
    for raw_data_dir in raw_data_dirs:
        prefix.append(raw_data_dir.split('\\')[-1].split("_")[0])

    try:
        # threads = []
        for result_file in result_files:
            if result_file.split('\\')[-1].split('_')[0] not in set(prefix):
                continue
            else:
                now_doing_msg(result_file, 'start', w2t)
                data_process(result_file, raw_data_dirs, av, rr, axis, vel, trq, w2t, rpm)
            # threads.append(Thread(target=data_process, args=(result_file, raw_data_dirs, AV, RR, RC, AXIS)))
        # [t.start() for t in threads]
        # [t.join() for t in threads]
    except Exception as Err:
        msg = f"出现错误：{Err}\n程序运行错误，请检查配置文件是否准确设定，以及数据文件组织是否正确，也有可能是结果文件损坏，尝试重新复制一份，再运行！"
        w2t(msg)

    w2t("----------------------------------------------------------")
    w2t("全部处理完毕")
    time_end = time()
    time_total = time_end - time_start
    msg = f"数据处理时间：{time_total // 3600:02.0f} h {time_total % 3600 // 60:02.0f} min {time_total % 60:02.0f} s"
    w2t(msg)


if __name__ == "__main__":
    stop = 0
    main(path=argv[1], av=argv[2], rr=argv[3], rpm=argv[4], axis=argv[5], vel=argv[6], trq=argv[7], w2t=argv[8])