SELECT `id` , `time` , `title`FROM `think_infor`WHERE `type` LIKE '$k-%' UNION ALL SELECT `id` , `time` , `title`FROM `think_infor2`WHERE `type` LIKE '$k-%' ORDER BY `time` DESCLIMIT 0,6
你带上表名不就行了?
不过你能得到 表1和表2的最新6条信息 吗?
SELECT A.id AS aid , A.time AS atime , A.title AS atitle, B.id AS bid , B.time AS btime , B.title AS btitle
FROM think_infor AS A, think_infor2 AS B
WHERE A.type LIKE '$k-%' AND B.type=A.type
ORDER BY A.time DESC LIMIT 0,6
SELECT A.id AS aid , A.time AS atime , A.title AS atitle, B.id AS bid , B.time AS btime , B.title AS btitle
FROM think_infor AS A, think_infor2 AS B
WHERE A.type LIKE '$k-%' AND B.type=A.type
ORDER……
这个差不多,但是A表的记录是会有重复记录
SELECT `id` , `time` , `title`, 1 as cpfromFROM `think_infor`WHERE `type` LIKE '$k-%' UNION ALL SELECT `id` , `time` , `title` , 2 as cpfromFROM `think_infor2`WHERE `type` LIKE '$k-%' ORDER BY `time` DESCLIMIT 0,6
这个语句写的不是很清楚
引用 2 楼 yiwusuo 的回复:SELECT A.id AS aid , A.time AS atime , A.title AS atitle, B.id AS bid , B.time AS btime , B.title AS btitle
FROM think_infor AS A, think_infor2 AS B
WHERE A.type LIKE……
ORDER BY A.time DESC 后面加上:having count(A.id)=1
不就没重复了吗
我自己都觉得乱了,我还是说一下我想要的东西吧
有A表,B表,结构相同,数据不同,我想同时模糊查询一个字段`type`,选取两表中符合条件的前6条记录(根据`time`排序)
你自己写的不就可以吗?
SELECT `id` , `time` , `title`, 'think_infor' as tbl_name FROM `think_infor` WHERE `type` LIKE '$k-%'UNIONSELECT `id` , `time` , `title`, 'think_infor2' FROM `think_infor2` WHERE `type` LIKE '$k-%'ORDER BY `time` DESCLIMIT 0,6
你自己写的不就可以吗?
SQL code
?
123456789
SELECT `id` , `time` , `title`, 'think_infor' as tbl_name FROM `think_infor` WHERE `type` LIKE '$k-%'UNIONSELECT `id` , `time` , `title`, 'think_in……
SELECT 不存在的字段这也可以
你是指 'think_infor' as tbl_name 这个吗?
那么你认为 SELECT `id` as iidd FROM `think_infor` 这样可以吗?
肯定可以了,不就是给个名字么
接下来就不说了