Пример #1
0
def visible(element):
    if element.parent.name in [
            'style', 'script', '[document]', 'head', 'title'
    ]:
        return False
    elif re.match('<!--.*-->', strip_non_ascii(element)):
        return False
    return True
Пример #2
0
def split_into_sentences(txt):
    txt = strip_non_ascii(txt)
    sents = map(clean_str, sent_tokenize(txt))
    return filter(lambda s: len(s.strip()) > 5, sents)
Пример #3
0
def split_into_sentences(txt):
    txt = strip_non_ascii(txt)
    sents = map(clean_str,sent_tokenize(txt))
    return filter(lambda s: len(s.strip()) > 5, sents)
Пример #4
0
def visible(element):
    if element.parent.name in ['style', 'script', '[document]', 'head', 'title']:
        return False
    elif re.match('<!--.*-->', strip_non_ascii(element)):
        return False
    return True