問題描述

我寫了一個處理文本的文件就是把文本中所有的符號都替換掉，替換成空格。用的python中maketrans和translate。其中在使用對于ASCII編碼的文件時是正常的，但對于utf-8文件時，就報錯，提示maketrans中的參數不等長，但是明明是一樣長的啊：

File '/Users/lgq/Desktop/p3.py', line 10, in text_to_words

'abcdefghijklmnopqrstuvwxyz ')

ValueError: the first two maketrans arguments must have equal length

我查了一下說是maketrans在utf-8下不能用，那我在utf-8下該怎么替換掉字符呢，求各位大神指點。

def text_to_words(the_text): ''' Return a list of words with all punctuation removed,and all in lowercase. ''' my_substitutions = the_text.maketrans(# If you find any of these'ABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789!'#$%&()*+,-./:;<=>?@[]^_`{|}~’',# Replace them by these'abcdefghijklmnopqrstuvwxyz ') # Translate the text now. cleaned_text = the_text.translate(my_substitutions) wds = cleaned_text.split() return wdsdef get_words_in_book(filename): ''' Read a book from filename, and return a list of its words.''' f = open(filename, 'r', encoding = 'utf-8') content = f.read() f.close() wds = text_to_words(content) return wdsbook_words = get_words_in_book('alice.txt')print('There are {0} words in the book, the first 100 aren{1}'.format(len(book_words), book_words[:100]))

問題解答

首先 這兩個字符串長度不相等， ' 是一個字符， 也是一個字符你可以用 len() 查看。然后關于字符串什么的問題，最好說明 python 的版本

maketrans 參數長度不相等

my_substitutions = the_text.maketrans(# If you find any of these'ABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789!'#$%&()*+,-./:;<=>?@[]^_`{|}~’',# Replace them by these'abcdefghijklmnopqrstuvwxyz ')

測試代碼：

from string import translate, maketransdef text_to_words(the_text): ''' Return a list of words with all punctuation removed,and all in lowercase. ''' my_substitutions = maketrans(# If you find any of these'ABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789!'#$%&()*+,-./:;<=>?@[]^_`{|}~’',# Replace them by these'abcdefghijklmnopqrstuvwxyz ') # Translate the text now. cleaned_text = the_text.translate(my_substitutions) wds = cleaned_text.split() return wdstext_to_words(’ABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789!'#$%&()*+,-./:;<=>?@[]^_`{|}~’測試’)

output

[’abcdefghijklmnopqrstuvwxyz’, ’xe6xb5x8bxe8xafx95’]

這是 python2 的運行結果