solr_match.py 文件源码

def parse_aiml_text(text):
    text = '<p>' + text + '</p>'
    soup = BeautifulSoup(text, 'lxml')
    tokens = []
    try:
        for c in soup.p.children:
            if isinstance(c, NavigableString):
                token = c.string.strip()
                if token:
                    tokens.append(token)
    except Exception as ex:
        logger.warn(ex)
        return text
    return ' '.join(tokens)