Question

我有一个包含一些列的表，其中一个包含我用来运行的时间戳目前有四个单一选择来计算SELECT count()并返回信息比如“超过1周的n行”，“超过2周的n行”，......

如何将四个查询转换为一个有希望的SQL语句跑得更快？

陈述如下：

SELECT count(foo_pk) AS oneweek FROM foo WHERE foo_timstamp < DATE_SUB(now(), INTERVAL 1 week)

Answer 1

我不是MySql人，但在SQL Server中我会像这样使用CASE语句：

SELECT
    SUM(CASE WHEN foo_timstamp < DATE_SUB(now(), INTERVAL 1 week) THEN 1 ELSE 0 END) as oneweek,
    SUM(CASE WHEN foo_timstamp < DATE_SUB(now(), INTERVAL 2 week) THEN 1 ELSE 0 END) as twoweek,
    SUM(CASE WHEN foo_timstamp < DATE_SUB(now(), INTERVAL 3 week) THEN 1 ELSE 0 END) as threeweek,
    SUM(CASE WHEN foo_timstamp < DATE_SUB(now(), INTERVAL 4 week) THEN 1 ELSE 0 END) as fourweek
FROM foo
WHERE 
    foo_timstamp < DATE_SUB(now(), INTERVAL 1 week)

似乎这个语句也可以在MySql中使用，所以虽然我的语法可能有些偏差，但我认为像上面的SQL这样的东西应该有效。

Answer 2

SELECT 
      COUNT(1) AS olderthanoneweek, 
      COUNT(CASE WHEN foo_timstamp < DATE_SUB(now(), INTERVAL 2 week) THEN 1 END) AS olderthantwoweek,
      COUNT(CASE WHEN foo_timstamp < DATE_SUB(now(), INTERVAL 3 week) THEN 1 END) AS olderthanthreeweek, 
      COUNT(CASE WHEN foo_timstamp < DATE_SUB(now(), INTERVAL 4 week) THEN 1 END) AS olderthanfourweek 
FROM foo 
WHERE foo_timstamp < DATE_SUB(now(), INTERVAL 1 week)

Answer 3

未尝试，但是..

select count(foo_pk), 
if(foo_timestamp<date_sub(now(),interval 1 week,"in the last week",
if(foo_timestamp<date_sub(now(),interval 2 week,"in the last 2 weeks",
if(foo_timestamp<date_sub(now(),interval 3 week,"in the last 3 weeks",
if(foo_timestamp<date_sub(now(),interval 4 week,"in the last 4 weeks",
                                                "4 weeks or more   ")))) as foo_label
order by foo_label
group by foo_label;

Answer 4

如果您愿意，可以使用这样的行：

SELECT DATE_SUB(now(), INTERVAL) AS WEEK_DIFF,
  count(foo_pk) AS COUNT_PER_WEEK
FROM foo
GROUP BY DATE_SUB(now(), INTERVAL);

Answer 5

select count(*), 
       round(datediff(curdate(), foo_pk) / 7)
         as week_diff
from wolffinder_experiment
group by week_diff
order by week_diff
having week_diff <= 4
limit 4;

SQL优化：将四个语句合二为一

5 个答案: