trigram.py 文件源码

def strip_punctuation(text):
    """
    strips the punctuation from a bunch of text
    """
    # build a translation table for string.translate:
    # there are other ways to do this:

    # create a translation table to replace all punctuation with spaces
    #    -- then split() will remove the extra spaces
    punctuation = string.punctuation
    punctuation = punctuation.replace("'", "")  # keep apostropies
    punctuation = punctuation.replace("-", "")  # keep hyphenated words
    # building a translation table
    table = {}
    for c in punctuation:
        table[ord(c)] = ' '
    # remove punctuation with the translation table
    text = text.translate(table)
    # remove "--" -- can't do multiple characters with translate
    text = text.replace("--", " ")

    return text