def test_spider(spider_type):
"""
test spider
"""
# 定义fetcher,parser和saver, 你也可以重写这三个类中的任何一个
fetcher = spider.Fetcher(normal_max_repeat=3, normal_sleep_time=0, critical_max_repeat=5, critical_sleep_time=5)
parser = MyParser(max_deep=1, max_repeat=2)
saver = spider.Saver(file_name="out_%s.txt" % spider_type)
# 定义Url过滤, UrlFilter使用Set, 适合Url数量不多的情况
black_patterns = (spider.CONFIG_URLPATTERN_FILES, r"binding", r"download", )
white_patterns = ("^http[s]{0,1}://(www\.){0,1}(wandoujia|(zhushou\.360))\.(com|cn)", )
url_filter = spider.UrlFilter(black_patterns=black_patterns, white_patterns=white_patterns, capacity=None)
# 确定使用ThreadPool还是ProcessPool
if spider_type == "thread":
web_spider = spider.WebSpiderT(fetcher, parser, saver, url_filter=url_filter, monitor_sleep_time=5)
else:
web_spider = spider.WebSpiderP(fetcher, parser, saver, url_filter=url_filter, monitor_sleep_time=5)
parser_num = 1 if spider_type == "thread" else 3
# 首先抓取一次豌豆荚页面,抓取完成之后不停止monitor
web_spider.set_start_url("http://www.wandoujia.com/apps", ("wandoujia",), priority=0, deep=0, critical=False)
web_spider.start_work_and_wait_done(fetcher_num=10, parser_num=parser_num, is_over=False)
# 然后抓取360应用商店页面,抓取完成之后停止monitor
web_spider.set_start_url("http://zhushou.360.cn/", ("360app",), priority=0, deep=0, critical=False)
web_spider.start_work_and_wait_done(fetcher_num=10, parser_num=parser_num, is_over=True)
return
def test_spider(mysql, spider_type):
"""
test spider
"""
# 定义fetcher, parser和saver, 你也可以重写这三个类中的任何一个
fetcher = spider.Fetcher(normal_max_repeat=3, normal_sleep_time=0, critical_max_repeat=5, critical_sleep_time=5)
# parser = spider.Parser(max_deep=1, max_repeat=2)
parser = MyParser(max_deep=1, max_repeat=2)
# 定义Url过滤
black_patterns = (spider.CONFIG_URLPATTERN_FILES, r"binding", r"download", )
white_patterns = ("^http[s]{0,1}://(www\.){0,1}(wandoujia|(zhushou\.360)|duba_\d)\.(com|cn)", )
if not mysql:
saver = spider.Saver(save_pipe=open("out.txt", "w", encoding="utf-8"))
# UrlFilter, 使用Set, 适合Url数量不多的情况
url_filter = spider.UrlFilter(black_patterns=black_patterns, white_patterns=white_patterns, capacity=None)
else:
saver = spider.SaverMysql(host="localhost", user="******", passwd="123456", database="default")
saver.change_sqlstr("insert into t_test(url, title, getdate) values (%s, %s, %s);")
# UrlFilter, 使用BloomFilter, 适合Url数量巨大的情况
url_filter = spider.UrlFilter(black_patterns=black_patterns, white_patterns=white_patterns, capacity=10000)
# 确定使用ThreadPool还是ProcessPool
if spider_type == "thread":
web_spider = spider.WebSpiderT(fetcher, parser, saver, url_filter=url_filter, monitor_sleep_time=5)
else:
web_spider = spider.WebSpiderP(fetcher, parser, saver, url_filter=url_filter, monitor_sleep_time=5)
parser_num = 1 if spider_type == "thread" else 3
# 首先抓取一次豌豆荚页面,抓取完成之后不停止monitor
web_spider.set_start_url("http://www.wandoujia.com/apps", ("wandoujia",), priority=0, deep=0, critical=False)
web_spider.start_work_and_wait_done(fetcher_num=10, parser_num=parser_num, is_over=False)
# 然后抓取360应用商店页面,并试验critical参数的作用,抓取完成之后停止monitor
web_spider.set_start_url("http://zhushou.360.cn/", ("360app",), priority=0, deep=0, critical=False)
for i in range(5):
web_spider.set_start_url("https://www.duba_%d.com/" % i, ("critical",), priority=0, deep=0, critical=True)
web_spider.start_work_and_wait_done(fetcher_num=10, parser_num=parser_num, is_over=True)
return
def item_save(self, url, keys, item):
"""
这里只需要重写item_save函数,参数含义及返回结果见框架
"""
if keys[0] == "Totals":
self.save_pipe_total.write("\t".join([keys[1]] + [str(i) for i in item]) + "\n")
elif keys[0] == "PerGame":
self.save_pipe_pergame.write("\t".join([keys[1]] + [str(i) for i in item]) + "\n")
else:
return False
return True
if __name__ == "__main__":
"""
main流程
"""
# 初始化fetcher, parser和saver
fetcher = NBAFetcher(critical_max_repeat=3, critical_sleep_time=0)
parser = NBAParser(max_deep=-1, max_repeat=3)
saver = NBASaver(file_name_total="nba_total.txt", file_name_pergame="nba_pergame.txt")
# 初始化爬虫, 并传入初始Url
nba_spider = spider.WebSpiderP(fetcher, parser, saver, url_filter=None)
nba_spider.set_start_url(url_player_index, ("index",), critical=True)
# 开启10个线程抓取数据
nba_spider.start_work_and_wait_done(fetcher_num=10, is_over=True)
exit()