' UCS-2'编解码器不能对61-61位的字符进行编码

时间:2017-08-16 13:43:11

标签: python json unicode encode

当我运行我的Python代码并打印(item)时,我收到以下错误:

UnicodeEncodeError: 'UCS-2' codec can't encode characters in position 61-61: Non-BMP character not supported in Tk

这是我的代码:

def getUserFollowers(self, usernameId, maxid = ''):
    if maxid == '':
        return self.SendRequest('friendships/'+ str(usernameId) +'/followers/?rank_token='+ self.rank_token,l=2)
    else:
        return self.SendRequest('friendships/'+ str(usernameId) +'/followers/?rank_token='+ self.rank_token + '&max_id='+ str(maxid))
def getTotalFollowers(self,usernameId):
    followers = []
    next_max_id = ''
    while 1:
        self.getUserFollowers(usernameId,next_max_id)
        temp = self.LastJson

        for item in temp["users"]:
            print(item)
            followers.append(item)

        if temp["big_list"] == False:
            return followers            
        next_max_id = temp["next_max_id"]

我该如何解决这个问题?

1 个答案:

答案 0 :(得分:1)

在不知道temp["users"]的内容的情况下难以猜测,但错误表明它包含非BMP unicode字符,例如emoji

如果您尝试在IDLE中显示该内容,则会立即出现此类错误。重现的简单示例(在ID 3.5 for Python 3.5上):

>>> t = "ab \U0001F600 cd"
>>> print(t)
Traceback (most recent call last):
  File "<pyshell#5>", line 1, in <module>
    print(t)
UnicodeEncodeError: 'UCS-2' codec can't encode characters in position 3-3: Non-BMP character not supported in Tk

\U0001F600表示unicode字符U + 1F600 笑脸

错误确实是由Tk不支持代码大于FFFF的unicode字符引起的。一个简单的解决方法是从字符串中过滤掉它们:

def BMP(s):
    return "".join((i if ord(i) < 10000 else '\ufffd' for i in s))

'\ufffd'是unicode U + FFFD REPLACEMENT CHARACTER 的Python表示。

我的例子变成:

>>> t = "ab \U0001F600 cd"
>>> print(BMP(t))
ab � cd

所以你的代码会变成:

for item in temp["users"]:
    print(BMP(item))
    followers.append(item)