rsc.py 文件源码

def parse_rsc_html(htmlstring):
    """Messy RSC HTML needs this special parser to fix problems before creating selector."""
    converted = UnicodeDammit(htmlstring)
    if not converted.unicode_markup:
        raise UnicodeDecodeError('Failed to detect encoding, tried [%s]')
    root = fromstring(htmlstring, parser=HTMLParser(recover=True, encoding=converted.original_encoding))
    # Add p.otherpara tags around orphan text
    newp = None
    for child in root.get_element_by_id('wrapper'):
        if newp is not None:
            if child.tag in BLOCK_ELEMENTS or child.get('id', '').startswith('sect') or child.getnext() is None:
                child.addprevious(newp)
                newp = None
            else:
                newp.append(child)
        if newp is None and child.tag in BLOCK_ELEMENTS and child.tail and child.tail.strip():
            newp = Element('p', **{'class': 'otherpara'})
            newp.text = child.tail
            child.tail = ''
    return root