movie_spider.py 文件源码

def parse_movie(self, response):
        content_info = response.css('div.contentinfo')
        movie = Movie()
        symbol1 = u'?'
        symbol2 = u'?'

        rex = r'%s(.*)%s' % (symbol1, symbol2)
        movie['title'] = content_info.css('h1 a::text').re_first(rex)

        # logging.log(logging.INFO, "parse_movie " + movie['title'])

        text = content_info.css('div#text')
        t_msg_font = text.css('div.t_msgfont')
        if len(t_msg_font) > 0:
            movie['cover'] = t_msg_font.css(' img::attr(src)').extract_first()
            movie['detail'] = self.parse_detail(t_msg_font.css('::text'))
        else:
            movie['cover'] = text.css(' p img::attr(src)').extract_first()
            movie['detail'] = self.parse_detail(text.css(' p::text'))
            thumbnails = text.css(' p img::attr(src)').extract()
            if movie['cover'] in thumbnails:
                thumbnails.remove(movie['cover'])
            movie['thumbnails'] = thumbnails
        download_links = text.css(' table tbody tr td a')
        download_links_array = []
        for link_item in download_links:
            download_link = DownloadLink()
            download_link['title'] = link_item.css('::text').extract_first()
            download_link['link'] = link_item.css('::attr(href)').extract_first();
            download_links_array.append(dict(download_link))
        movie['download_links'] = download_links_array
        return movie