（名称错误：未定义全局名称）将列表传递给函数时出错

Question

我很抱歉发布了一个很长的代码。我提取了生成和使用列表lower_lip_under_upper_list_plist的代码部分。链接使用此列表的函数时出错。generate_t_u()函数在time()函数中被调用，但是time()函数不会识别这个列表，因为它没有被传递到那里。但是如果我在这个函数中传递它，那么我也必须对main进行更改。同样，我也会得到同样的错误。在

def time(transcriptionFile) :
    """ This function changes the time values in secs from the transcription file and keeps a list of start time for each phoneme. """
    with open("transcriptions.txt", "r") as tFile :
        timeList = []
        t_u = 0.0
        for line in tFile :
            li = line.split()
            if li :
                start_time = (int(li[0]) / 10000000.)
                timeList.append(start_time)
                #print timeList 

    generate_t_u(timeList, lower_lip_under_upper_list_plist)

def generate_t_u(timeList, lower_lip_under_upper_list_plist)
    """ It generates the regular time intervals t(u) values for sampling. """
    i = 0        
    while i < len(timeList) - 1 :
        # if the number is in the range
        # do the calculations and move to the next number
        if t_u > timeList[i] and t_u < timeList[i + 1] :
            #print "\n The t_u value:", t_u, 'is between',
            #print "start:", timeList[i], " and end: ", timeList[i+1]
            poly = poly_coeff(timeList[i], timeList[i + 1], t_u)
            Newton(poly, lower_lip_under_upper_list_plist[i],  lower_lip_under_upper_list_plist[i + 1])
            t_u = t_u + 0.04 # regular time interval

        # if the number is at the lower boundary of the range, no need of calculation as u = 0
        elif t_u == timeList[i] :
            #print "\n The t_u value:", t_u, 'is on the boundary of',
            #print "start:", timeList[i], " and end: ", timeList[i+1]
            #print "u : 0"
            lower_lip_under_upper_list_bezier(0, lower_lip_under_upper_list_plist[i], lower_lip_under_upper_list_plist[i + 1])
            t_u = t_u + 0.04 # regular time interval

        # if the number is at the upper boundary of the range, no need of calculation as u = 1
        elif t_u == timeList[i + 1] :
            #print "\n The t_u value:", t_u, 'is on the boundary of',
            #print "start:", timeList[i], " and end: ", timeList[i+1]
            #print " u : 1"
            lower_lip_under_upper_list_bezier(1, lower_lip_under_upper_list_plist[i], lower_lip_under_upper_list_plist[i + 1])
            t_u = t_u + 0.04 # regular time interval

        # if the number isn't in the range, move to the next range
        else :
            i += 1

def Newton(poly, p0, p3) :
    """ Newton's method for finding the root of a polynomial. Here the root is the 'u' value"""
    poly_diff = poly_differentiate(poly)
    counter = 0
    epsilon = 0.000000000001
    x = 0.5 # initial guess value

    while True:
        x_n = x - (float(poly_substitute(poly, x)) / poly_substitute(poly_diff, x))
        counter += 1
        if abs(x_n - x) < epsilon :
            break
        x = x_n
        #print "\tIteration " , counter , " : ", x_n

     print "u:", (x_n)

    lower_lip_under_upper_list_bezier(x_n, p0, p3)      

def param_lists_separate(pList) :
    """ Separating the parameter values of each feature into individual lists """
    v = [[inner[1][i] for outer in pList for inner in outer]
        for i in range(len(pList[0][0][1]))]

    lower_lip_under_upper_list_plist = v[0]
    lips_part_plist = v[1]
    lips_spread_plist = v[2]
    jaw_open_plist = v[3]
    lips_round_plist = v[4]  

    return lower_lip_under_upper_list_plist

def lower_lip_under_upper_list_bezier(x_n, p0, p3) :
    """ Calculating sampling points using rational bezier curve equation"""
    u = x_n
    p1 = p0
    p2 = p3

    lower_lip_under_upper_list_p_u = math.pow(1 - u, 3) * p0 + 3 * u * math.pow(1 - u, 2) * p1 \
                                 + 3 * (1 - u) * math.pow(u, 2) * p2 + math.pow(u, 3) * p3
    lower_lip_under_upper_list_p_u = lower_lip_under_upper_list_p_u * w
    d = math.pow(1 - u, 3) * w + 3 * u * w * math.pow(1 - u, 2) + 3 * (1 - u) * w * math.pow(u, 2) + math.pow(u, 3) * w
    lower_lip_under_upper_list_p_u = lower_lip_under_upper_list_p_u / d

    print "\n p(u) values for the feature lower lip under upper list \n"
    print "p(u): ", lower_lip_under_upper_list_p_u
    return lower_lip_under_upper_list_p_u

if __name__ == "__main__" :
    time("transcriptions.txt") 

错误是：

NameError: global name 'lower_lip_under_upper_list_plist' is not defined.

此错误位于time()函数的lower_lip_under_upper_list_bezier(0, lower_lip_under_upper_list_plist[i], lower_lip_under_upper_list_plist[i + 1])行。在

Answer 1

考虑time函数中的这一行：

def time(transcriptionFile) :
    ...
    generate_t_u(timeList, lower_lip_under_upper_list_plist)

lower_lip_under_upper_list_plist未在time函数中定义，因此Python在全局范围内查找它。它也没有在那里定义，因此Python引发了一个NameError，表示变量{}没有定义。在

---

顺便说一句，回溯错误消息包含一个行号，指示是什么行导致了NameError。这将有助于您（和我们）集中精力解决问题，包括行号。在

Answer 2

lower_lip_under_upper_list_plist应该是generate_t_u的返回值吗？尝试将generate_t_u更改为：

def generate_t_u(time_list):
    lower_lip_under_upper_list_plist = []
    # ... the function remains the same
    return lower_lip_upper_list_plist

我不确定param_lists_separate在哪里被调用，但看起来其中的绑定也应该是全局的。在generate_t_u中也存在类似的问题，因为t_u没有绑定在那里，如果它引用time中的t_u，那么它可能应该作为参数传入。如果有许多相关的绑定和方法作用于数据，那么您可能应该考虑创建一个将数据和操作绑定在一起的类。在

考虑做一些类似以下的事情：

^{pr2}$

面向对象编程最基本的原则之一是将算法与它们处理和/或需要的数据捆绑在一起。你可以在这里应用这个想法。您的全局绑定将成为对象的属性，因此它们始终可用作实例方法中self的属性。您可能已经注意到，不再需要传递来自lower_lip_under_upper_list_plist之类的值。您只需将索引传递到列表中，因为列表本身是对象的属性。在

将您的算法重构为class应该可以消除所有地方对全局变量的需要。它还将最小化参数的数量，并让您集中精力以一种干净和模块化的方式实现算法。在