html_to_telegraph.py 文件源码

def _fragments_from_string(html_string):
    fragments = html.fragments_fromstring(html_string)
    if not len(fragments):
        return []
    # convert and append text node before starting tag
    if not isinstance(fragments[0], html.HtmlElement):
        if len(fragments[0].strip()) > 0:
            if len(fragments) == 1:
                return html.fragments_fromstring('<p>%s</p>' % fragments[0])
            else:
                paragraph = _create_element('p')
                paragraph.text = fragments[0]
                fragments[1].addprevious(paragraph)
                fragments.insert(1, paragraph)

        fragments.pop(0)
        if not len(fragments):
            return []

    # remove xml instructions (if cleaning is disabled)
    for instruction in fragments[0].xpath('//processing-instruction()'):
        instruction.drop_tag()

    return fragments