def _gen_item(self, element): item = Article() item.url = self.url item.title = element.find('h4').text() item.author = u'廖雪峰'#element.find('').text() item.category = 'Dev' item.tags = 'Python' contentWrapper = element.find('.x-wiki-content')#.remove('.postComment') item.summary = tool.Tool().replace2(contentWrapper.outerHtml().encode('utf-8')) item.crawl_time = datetime.datetime.now() return item