import matplotlib.pyplot as plt
import numpy as np
import os
import pandas as pd
import re
import shutil

def merge(x1,x2):
    x = []
    for i, j in zip(x1,x2):
        x.append(i)
        x.append(j)
    x.append(x1[-1])
    return np.asarray(x)

def move(src, dst):
    shutil.move(src, dst)

def cal_mid(a):
    b = []
    for i in range(len(a) - 1):
        b.append((a[i] + a[i + 1]) / 2)
    return np.asarray(b)


# 将.dat文件转化为.csv文件，并输出点数低于阈值的文件名
def prc_svcsv(read_path, save_path):
    # 去除异常文件
    if "n642415" in read_path:
        print("remove: ", read_path)
        return
    elif "naca23015" in read_path:
        print("remove: ", read_path)
        return
    elif "naca23018" in read_path:
        print("remove: ", read_path)
        return
    elif "naca2412" in read_path:
        print("remove: ", read_path)
        return
    elif "naca1.dat" in read_path:
        print("remove: ", read_path)
        return
    elif "e664ex" in read_path:
        print("remove: ", read_path)
        return
    elif "lrn1007" in read_path: # 科学计数法特殊处理
        print("remove: ", read_path)
        return
    elif "naca64a010" in read_path:
        print("remove: ", read_path)
        return
    elif "s1221" in read_path:
        print("remove: ", read_path)
        return
    elif "sc1095r8" in read_path:
        print("remove: ", read_path)
        return

    file = open(read_path, mode='r')

    strs = file.readlines()

    dict = {}
    x = []
    y = []

    for tmp_str in strs:
        if tmp_str == "":
            continue

        tmp_str = tmp_str.strip()
        while "  " in tmp_str:
            tmp_str = tmp_str.replace("  ", " ")

        tmp_strs = tmp_str.split(" ")
        if "	" in tmp_str:
            tmp_strs = tmp_str.split("	")

        if re.match(r"^([-,0-9]{0,}[.][0-9]*)$", tmp_strs[0]) == None:
            continue
        if re.match(r"^([-,0-9]{0,}[.][0-9]*)$", tmp_strs[1]) == None:
            continue

        # 去除异常值
        if float(tmp_strs[1]) > 10:
            continue

        x.append(float(tmp_strs[0]))
        y.append(float(tmp_strs[1]))

    df = pd.DataFrame(dict)
    df['x'] = x
    df['y'] = y
    df.to_csv(save_path, index=False)

    # 输出点数少于阈值的文件名及点数
    # if df.shape[0] <= 17:
    #     print(read_path + "\t" + str(df.shape[0]))

# rename: from "*.DAT" to "*.dat"
def file_rename(file_name):
    os.rename(file_name, file_name[:-3]+"dat")

# count
def csv_count(file_path):
    df = pd.read_csv(file_path)
    # if len(df['x']) <= 17:
    #     print(file_path, len(df['x']))
    return len(df['x'])

# find the number out of range and modify
def csv_out_range(file_path):
    df = pd.read_csv(file_path)

    # test coordinate of x
    list_x = df['x']

    for i, j in enumerate(list_x):
        if j > 1:
            list_x[i] = 1
            print(file_path, i, j)
        elif j < 0:
            list_x[i] = 0
            print(file_path, i, j)

    df.to_csv(file_path, index=False)

    # test coordinate of y
    list_y = df['y']

    for i, j in enumerate(list_y):
        if j > 0.45:
            # list_y[i] = 1
            print(file_path, i, j)
        elif j < -0.45:
            # list_y[i] = -1
            print(file_path, i, j)

    # df.to_csv("test1.csv", index=False)


def csi(X1, Y1, X2, Y2, file_name):
    '''
        三次样条插值
    :param X1:
    :param Y1:
    :param X2:
    :param Y2:
    :param file_name:
    :return:
    '''
    plt.rcParams['font.sans-serif'] = ['SimHei']  # 用来正常显示中文标签
    plt.rcParams['axes.unicode_minus'] = False  # 用来正常显示负号

    # print(X1)
    # print(Y1)
    # print(X2)
    # print(Y2)

    # new_x = np.arange(0, 1.01, 0.01)  # 定义差值点
    new_x_u = cal_mid(X1)
    new_x_d = cal_mid(X2)


    # 进行样条差值
    import scipy.interpolate as spi

    # 进行三次样条拟合
    ipo3_u = spi.splrep(X1, Y1, k=3)  # 样本点导入，生成参数
    iy3_u = spi.splev(new_x_u, ipo3_u)  # 根据观测点和样条参数，生成插值
    ipo3_d = spi.splrep(X2, Y2, k=3)  # 样本点导入，生成参数
    iy3_d = spi.splev(new_x_d, ipo3_d)  # 根据观测点和样条参数，生成插值

    # 作图
    plt.plot(X1, Y1, 'o', label='样本点上')
    plt.plot(new_x_u, iy3_u, '*', label='插值点上')
    plt.plot(X2, Y2, 'o', label='样本点下')
    plt.plot(new_x_d, iy3_d, '*', label='插值点下')
    plt.ylim(-0.3, 0.3)
    plt.ylabel('指数')
    plt.title('机翼数据三次样条插值拟合结果')
    plt.legend()

    plt.show()

    # 将数据合并，最后保存为csv文件
    x_up = merge(X1[::-1], new_x_u[::-1])
    y_up = merge(Y1[::-1], iy3_u[::-1])
    x_down = merge(X2, new_x_d)
    y_down = merge(Y2, iy3_d)
    x = np.concatenate((x_up, x_down))
    y = np.concatenate((y_up, y_down))
    dict = {'x': x, 'y': y}
    df = pd.DataFrame(dict)
    df.to_csv(file_name, index=False)


def hermite(xi, yi):
    from scipy.interpolate import KroghInterpolator

    x = np.linspace(0, 1, 20)

    # xi = np.array([0, 0, 1, 1])
    # yi = np.array([1, 0, 2, 3])
    interpolant = KroghInterpolator(xi, yi)
    plt.figure()
    plt.plot(x, interpolant(x), 'r--', label='Hermite Interpolation')
    plt.plot(xi, yi, 'go', label='nodes', markersize=8)
    plt.legend(loc=9)
    plt.xlim(0, 1)
    plt.title('$埃尔米特插值$')
    plt.show()

def lagrange(x, y):
    plt.rcParams['font.sans-serif'] = ['SimHei']  # 用来正常显示中文标签
    plt.rcParams['axes.unicode_minus'] = False  # 用来正常显示负号

    from scipy.interpolate import lagrange
    # x = [1, 2, 3, 4]
    # y = [4, 15, 40, 85]
    lag_ret = lagrange(x,y)

    x0 = np.arange(0, 1, 0.02)

    plt.figure()
    plt.plot(x0, lag_ret(x0), label="Lagrange Interpolation")
    plt.plot(x, y, 'go', label="notes", markersize=8)
    plt.legend(loc=9)
    plt.xlim(0, 1)
    plt.ylim(-0.3, 0.3)
    plt.title('$拉格朗日插值$')
    plt.show()

def Monotonic(path):
    df = pd.read_csv(path)

    x = list(df['x'])

    state_change_count = 0
    state = x[0] > x[1] # True表示递减;False表示递增
    for i in range(2, len(x)):
        tmp_state = x[i-1] > x[i]
        if tmp_state != state:
            state_change_count += 1
            print("{}: {}->{}".format(i, state, tmp_state))
        state = tmp_state
    return state_change_count