tag.py 文件源码

def __call__(self, doc, encoding='UTF-8'):
        if isinstance(doc,
                      (str, bytes)):
            doc = fromstring(bytes(bytearray(doc,
                                             encoding=encoding)),
                             parser=HTMLParser(encoding=encoding))
        if not isinstance(doc,
                          HtmlElement):
            return None
        for cls in self.EXTRACTORS:
            extract = cls()
            tags_ = extract(doc)
            if tags_:
                tags = []
                for idx, tag in enumerate(tags_):
                    if idx < 2 and len(tag) > 16:
                        break
                    elif len(tag) < 16:
                        tags.append(tag)
                else:
                    if tags:
                        logger.info('TagExtractor got tags %s',
                                    tags)
                    return tags