crawl.py 文件源码

def fetch_html(cls, session, url_tuples):
        """Actual method to do fetch html action.

        Parameters
        ----------
            session : object
                a SQLAlchemy session object.
            url_tuples : list
                a list of url tuple (id, raw, status_code).
        """
        settings = Settings(cls.conf['crawl']['scrapy'])
        settings.set('ITEM_PIPELINES',
                     {'hoaxy.crawl.pipelines.HtmlPipeline': 300})
        process = CrawlerProcess(settings)
        sll = cls.conf['logging']['loggers']['scrapy']['level']
        logging.getLogger('scrapy').setLevel(logging.getLevelName(sll))
        logger.warning('Number of url to fetch html is: %s', len(url_tuples))
        process.crawl(
            HtmlSpider,
            session=session,
            url_tuples=url_tuples,
            excluded_domains=cls.conf['crawl']['excluded_domains'])
        process.start()