def crawl_weibo_datas(uid): limit = get_max_home_page() cur_page = 1 while cur_page <= limit: url = home_url.format(uid, cur_page) html = get_page(url) weibo_datas = get_wbdata_fromweb(html) if not weibo_datas: crawler.warning("user {} has no weibo".format(uid)) return insert_weibo_datas(weibo_datas) domain = public.get_userdomain(html) cur_time = int(time.time()*1000) ajax_url_0 = ajax_url.format(domain, 0, domain, uid, cur_page, cur_page, cur_time) ajax_url_1 = ajax_url.format(domain, 1, domain, uid, cur_page, cur_page, cur_time+100) if cur_page == 1: # here we use local call to get total page number total_page = get_total_page(crawl_ajax_page(ajax_url_1)) if total_page < limit: limit = total_page cur_page += 1 app.send_task('tasks.home.crawl_ajax_page', args=(ajax_url_0,), queue='ajax_home_crawler', routing_key='ajax_home_info') app.send_task('tasks.home.crawl_ajax_page', args=(ajax_url_1,), queue='ajax_home_crawler', routing_key='ajax_home_info') set_seed_home_crawled(uid)
def crawl_weibo_datas(uid): limit = get_max_home_page() cur_page = 1 while cur_page <= limit: url = home_url.format(uid, cur_page) html = get_page(url) weibo_datas = get_wbdata_fromweb(html) if not weibo_datas: crawler.warning('用户id为{}的用户主页微博数据未采集成功,请检查原因'.format(uid)) return insert_weibo_datas(weibo_datas) domain = public.get_userdomain(html) ajax_url_0 = ajax_url.format(domain, 0, domain, uid, cur_page, cur_page) ajax_url_1 = ajax_url.format(domain, 1, domain, uid, cur_page, cur_page) if cur_page == 1: total_page = get_total_page(crawl_ajax_page(ajax_url_1)) if total_page < limit: limit = total_page cur_page += 1 app.send_task('tasks.home.crawl_ajax_page', args=(ajax_url_0,), queue='ajax_home_crawler', routing_key='ajax_home_info') app.send_task('tasks.home.crawl_ajax_page', args=(ajax_url_1,), queue='ajax_home_crawler', routing_key='ajax_home_info')
def test_get_total_home_page(self): """ 测试获取主页页数 :return: """ from page_parse import home with open('tests/asyncpersonhome.html', encoding='utf-8') as f: html = f.read() num = home.get_total_page(html) self.assertEqual(num, 18)
def crawl_weibo_datas(uid): limit = get_max_home_page() cur_page = 1 while cur_page <= limit: url = HOME_URL.format(uid, cur_page) if cur_page == 1: html = get_page(url, auth_level=1) else: html = get_page(url, auth_level=2) weibo_datas = get_data(html) if not weibo_datas: crawler.warning("user {} has no weibo".format(uid)) return # Check whether weibo created after time in spider.yaml timeafter = time.mktime( time.strptime(get_time_after(), '%Y-%m-%d %H:%M:%S')) length_weibo_datas = len(weibo_datas) for i in range(0, len(weibo_datas)): weibo_time = time.mktime( time.strptime(weibo_datas[i].create_time, '%Y-%m-%d %H:%M')) if weibo_time < timeafter: weibo_datas = weibo_datas[0:i] break WbDataOper.add_all(weibo_datas) # If the weibo isn't created after the given time, jump out the loop if i != length_weibo_datas - 1: break domain = public.get_userdomain(html) cur_time = int(time.time()*1000) ajax_url_0 = AJAX_URL.format(domain, 0, domain, uid, cur_page, cur_page, cur_time) ajax_url_1 = AJAX_URL.format(domain, 1, domain, uid, cur_page, cur_page, cur_time+100) if cur_page == 1: # here we use local call to get total page number total_page = get_total_page(crawl_ajax_page(ajax_url_1, 2)) auth_level = 1 else: auth_level = 2 if total_page < limit: limit = total_page app.send_task('tasks.home.crawl_ajax_page', args=(ajax_url_0, auth_level), queue='ajax_home_crawler', routing_key='ajax_home_info') app.send_task('tasks.home.crawl_ajax_page', args=(ajax_url_1, auth_level), queue='ajax_home_crawler', routing_key='ajax_home_info') cur_page += 1 SeedidsOper.set_seed_home_crawled(uid)
def test_get_total_home_page(self): """ 测试获取主页页数 :return: """ from page_parse import home url = TEST_SERVER + 'asyncpersonhome.html' resp = requests.get(url) resp.encoding = 'utf-8' html = resp.text num = home.get_total_page(html) self.assertEqual(num, 18)
def crawl_weibo_datas(uid): limit = get_max_home_page() cur_page = 1 while cur_page <= limit: url = home_url.format(uid, cur_page) html = get_page(url) weibo_datas = get_wbdata_fromweb(html) if not weibo_datas: crawler.warning("user {} has no weibo".format(uid)) return insert_weibo_datas(weibo_datas) domain = public.get_userdomain(html) page_id = public.get_pageid(html) cur_time = int(time.time() * 1000) ajax_url_0 = ajax_url.format(domain, 0, page_id, uid, cur_page, cur_page, cur_time) ajax_url_1 = ajax_url.format(domain, 1, page_id, uid, cur_page, cur_page, cur_time + 100) if cur_page == 1: # here we use local call to get total page number total_page = get_total_page(crawl_ajax_page(ajax_url_1)) if total_page < limit: limit = total_page cur_page += 1 app.send_task('tasks.home.crawl_ajax_page', args=(ajax_url_0, ), queue='ajax_home_crawler', routing_key='ajax_home_info') app.send_task('tasks.home.crawl_ajax_page', args=(ajax_url_1, ), queue='ajax_home_crawler', routing_key='ajax_home_info') set_seed_home_crawled(uid)
def crawl_weibo_datas(uid): limit = get_max_home_page() cur_page = 1 while cur_page <= limit: url = HOME_URL.format(uid, cur_page) if cur_page == 1: html = get_page(url, auth_level=1) else: html = get_page(url, auth_level=2) weibo_datas = get_data(html) if not weibo_datas: crawler.warning("user {} has no weibo".format(uid)) return WbDataOper.add_all(weibo_datas) domain = public.get_userdomain(html) cur_time = int(time.time()*1000) ajax_url_0 = AJAX_URL.format(domain, 0, domain, uid, cur_page, cur_page, cur_time) ajax_url_1 = AJAX_URL.format(domain, 1, domain, uid, cur_page, cur_page, cur_time+100) if cur_page == 1: # here we use local call to get total page number total_page = get_total_page(crawl_ajax_page(ajax_url_1, 2)) auth_level = 1 else: auth_level = 2 if total_page < limit: limit = total_page app.send_task('tasks.home.crawl_ajax_page', args=(ajax_url_0, auth_level), queue='ajax_home_crawler', routing_key='ajax_home_info') app.send_task('tasks.home.crawl_ajax_page', args=(ajax_url_1, auth_level), queue='ajax_home_crawler', routing_key='ajax_home_info') cur_page += 1 SeedidsOper.set_seed_home_crawled(uid)
def crawl_weibo_datas(uid): limit = get_max_home_page() cur_page = 1 while cur_page <= limit: url = HOME_URL.format(uid, cur_page) if cur_page == 1: html = get_page(url, auth_level=1) else: html = get_page(url, auth_level=2) weibo_datas = get_data(html) if not weibo_datas: crawler.warning("user {} has no weibo".format(uid)) return # Check whether weibo created after time in spider.yaml length_weibo_datas = len(weibo_datas) timeafter = time.mktime( time.strptime(get_time_after(), '%Y-%m-%d %H:%M:%S')) weibo_datas = [ weibo_datum for weibo_datum in weibo_datas if determine(weibo_datum, timeafter) ] WbDataOper.add_all(weibo_datas) # If the weibo isn't created after the given time, jump out the loop if len(weibo_datas) != length_weibo_datas: break domain = public.get_userdomain(html) cur_time = int(time.time() * 1000) ajax_url_0 = AJAX_URL.format(domain, 0, domain, uid, cur_page, cur_page, cur_time) ajax_url_1 = AJAX_URL.format(domain, 1, domain, uid, cur_page, cur_page, cur_time + 100) if cur_page == 1: # here we use local call to get total page number total_page = get_total_page(crawl_ajax_page(ajax_url_1, 2)) auth_level = 1 if total_page < limit: limit = total_page # Since the second ajax of page 1 has already been crawled # in the code above and has been stored in databse, # we only have to crawl the first ajax of page 1 crawl_ajax_page(ajax_url_0, auth_level) else: auth_level = 2 # Still the same as before # if total_page != limit: # limit = total_page # crawler.warning("total pagenum is {}".format(total_page)) crawl_ajax_page(ajax_url_0, auth_level) crawl_ajax_page(ajax_url_1, auth_level) cur_page += 1 SeedidsOper.set_seed_home_crawled(uid)
def crawl_weibo_datas(uid): limit = get_max_home_page() cur_page = 1 while cur_page <= limit: url = HOME_URL.format(uid, cur_page) if cur_page == 1: html = get_page(url, auth_level=1) else: html = get_page(url, auth_level=2) weibo_datas = get_data(html) if not weibo_datas: crawler.warning("user {} has no weibo".format(uid)) return # Check whether weibo created after time in spider.yaml length_weibo_datas = len(weibo_datas) timeafter = time.mktime( time.strptime(get_time_after(), '%Y-%m-%d %H:%M:%S')) weibo_datas = [ weibo_datum for weibo_datum in weibo_datas if determine(weibo_datum, timeafter) ] WbDataOper.add_all(weibo_datas) # If the weibo isn't created after the given time, jump out the loop if len(weibo_datas) != length_weibo_datas: break domain = public.get_userdomain(html) cur_time = int(time.time()*1000) ajax_url_0 = AJAX_URL.format(domain, 0, domain, uid, cur_page, cur_page, cur_time) ajax_url_1 = AJAX_URL.format(domain, 1, domain, uid, cur_page, cur_page, cur_time+100) if cur_page == 1: # here we use local call to get total page number total_page = get_total_page(crawl_ajax_page(ajax_url_1, 2)) auth_level = 1 if total_page < limit: limit = total_page # Since the second ajax of page 1 has already been crawled # in the code above and has been stored in databse, # we only have to crawl the first ajax of page 1 crawl_ajax_page(ajax_url_0, auth_level) else: auth_level = 2 # Still the same as before # if total_page != limit: # limit = total_page # crawler.warning("total pagenum is {}".format(total_page)) crawl_ajax_page(ajax_url_0, auth_level) crawl_ajax_page(ajax_url_1, auth_level) cur_page += 1 SeedidsOper.set_seed_home_crawled(uid)
def crawl_weibo_datas(uid): limit = get_max_home_page() cur_page = 1 # 自定义最大爬取的页数 max_page = 10 # end while cur_page <= limit: # 有些微博账号的照片多达两三千张,如果全部爬取比较浪费时间,这里先简单粗暴地根据当前微博的页数 # 进行限制。经过调查发现10页左右应该是比较理想的数字,电脑版微博一页有45条微博,那么一个账户就是 # 450条微博。 if cur_page > max_page: break # end url = home_url.format(uid, cur_page) html = get_page(url) domain = public.get_userdomain(html) # 只爬取微博个人用户的相片,如果是非个人用户(如政府,组织等)不爬取。 if domain not in ['103505', '100306', '100505', '']: set_seed_home_crawled(uid, 2) return # end weibo_datas, weibo_pics = get_wbdata_fromweb(html) if not weibo_datas: crawler.warning('用户id为{}的用户主页微博数据未采集成功,请检查原因'.format(uid)) return insert_weibo_datas(weibo_datas) # 如果非空,则将weibo_pics插入数据库中 if weibo_pics: insert_weibo_pics(weibo_pics) # end cur_time = int(time.time() * 1000) ajax_url_0 = ajax_url.format(domain, 0, domain, uid, cur_page, cur_page, cur_time) ajax_url_1 = ajax_url.format(domain, 1, domain, uid, cur_page, cur_page, cur_time + 100) if cur_page == 1: total_page = get_total_page(crawl_ajax_page(ajax_url_1)) if total_page < limit: limit = total_page cur_page += 1 app.send_task('tasks.home.crawl_ajax_page', args=(ajax_url_0, ), queue='ajax_home_crawler', routing_key='ajax_home_info') app.send_task('tasks.home.crawl_ajax_page', args=(ajax_url_1, ), queue='ajax_home_crawler', routing_key='ajax_home_info') # 在遍历完所有页数之后,将flag置位。放在这里表示所有页面都遍历过,不保证遍历成功后置位。可能以后还要优化,即在 # 某个回调函数中使用它。 set_seed_home_crawled(uid)