? 下面是摇奖概率调整后，一条实用的SQL统计查看结果：
?
? ?该SQL非常高效，绝对实用，不用group by 关键是用到case这个函数。
?
SELECT '2015-04-27 老的摇奖概率 1-11点数据', SUM(CASE t.result WHEN 1 THEN 1 ELSE 1 END) AS 总摇奖人数, SUM(CASE t.result WHEN 1 THEN 1 ELSE 0 END) AS 一等奖, SUM(CASE t.result WHEN 2 THEN 1 ELSE 0 END) AS 二等奖, SUM(CASE t.result WHEN 3 THEN 1 ELSE 0 END) AS 三等奖, SUM(CASE t.result WHEN 0 THEN 1 ELSE 0 END) AS 未中奖FROM mms.mobile_ernie t where ernie_time>='2015-04-27 00:00:00' and ernie_time<='2015-04-27 11:00:00'union allSELECT '2015-04-28 摇奖概率调整后 1-11点数据', SUM(CASE t.result WHEN 1 THEN 1 ELSE 1 END) AS 总摇奖人数, SUM(CASE t.result WHEN 1 THEN 1 ELSE 0 END) AS 一等奖, SUM(CASE t.result WHEN 2 THEN 1 ELSE 0 END) AS 二等奖, SUM(CASE t.result WHEN 3 THEN 1 ELSE 0 END) AS 三等奖, SUM(CASE t.result WHEN 0 THEN 1 ELSE 0 END) AS 未中奖FROM mms.mobile_ernie t where ernie_time>='2015-04-28 00:00:00' and ernie_time<='2015-04-28 11:00:00'
?
查询结果截图如下：
?
?