crawl.py 文件源码

def fetch(self, url, max_redirect):
        tries = 0
        exception = None
        while tries < self.max_tries:
            try:
                response = await self.session.get(
                    url, allow_redirects=False)
                break
            except aiohttp.ClientError as client_error:
                exception = client_error

            tries += 1
        else:
            self.record_statistic(FetchStatistic(url=url,
                                                 next_url=None,
                                                 status=None,
                                                 exception=exception,
                                                 size=0,
                                                 content_type=None,
                                                 encoding=None,
                                                 num_urls=0,
                                                 num_new_urls=0))
            return

        try:
            if is_redirect(response):
                location = response.headers['location']
                next_url = urllib.parse.urljoin(url, location)
                self.record_statistic(FetchStatistic(url=url,
                                                     next_url=next_url,
                                                     status=response.status,
                                                     exception=None,
                                                     size=0,
                                                     content_type=None,
                                                     encoding=None,
                                                     num_urls=0,
                                                     num_new_urls=0))

                if next_url in self.seen_urls:
                    return
                if max_redirect > 0:
                    self.add_url(next_url, max_redirect - 1)
                else:
                    print('redirect limit reached for %r from %r',
                          next_url, url)
            else:
                stat, links = await self.parse_links(response)
                self.record_statistic(stat)
                for link in links.difference(self.seen_urls):
                    self.q.put_nowait((link, self.max_redirect))
                self.seen_urls.update(links)
        finally:
            await response.release()