araby.py 文件源码

def strip_lastharaka(text):
    """Strip the last Haraka from arabic word except Shadda.
    The striped marks are :
        - FATHA,  DAMMA,  KASRA
        - SUKUN
        - FATHATAN,  DAMMATAN,  KASRATAN,  ,  ,  .
    Example:
        >>> text = u"?????????????"
        >>> stripTashkeel(text)
        ????????????
    @param text: arabic text.
    @type text: unicode.
    @return: return a striped text.
    @rtype: unicode.
    """
    if text:
        if is_vocalized(text):
            return  re.sub(LASTHARAKA_PATTERN, u'', text)
    return text