test_htmlparser.py 文件源码

def test_cdata_with_closing_tags(self):
        # see issue #13358
        # make sure that HTMLParser calls handle_data only once for each CDATA.
        # The normal event collector normalizes the events in get_events,
        # so we override it to return the original list of events.
        class Collector(EventCollector):
            def get_events(self):
                return self.events

        content = """<!-- not a comment --> &not-an-entity-ref;
                  <a href="" /> </p><p> &amp; <span></span></style>
                  '</script' + '>' </html> </head> </scripter>!"""
        for element in [' script', 'script ', ' script ',
                        '\nscript', 'script\n', '\nscript\n']:
            s = u'<script>{content}</{element}>'.format(element=element,
                                                        content=content)
            self._run_check(s, [("starttag", "script", []),
                                ("data", content),
                                ("endtag", "script")],
                            collector=Collector)