mysql多表聯(lián)合查詢操作實(shí)例分析

更新時(shí)間：2019年04月29日 09:26:06 作者：雨落知音

這篇文章主要介紹了mysql多表聯(lián)合查詢操作,結(jié)合實(shí)例形式分析了mysql多表聯(lián)合查詢的語(yǔ)法、功能、相關(guān)操作技巧與注意事項(xiàng),需要的朋友可以參考下

本文實(shí)例講述了mysql多表聯(lián)合查詢操作。分享給大家供大家參考，具體如下：

MySQL多表聯(lián)合查詢是MySQL數(shù)據(jù)庫(kù)的一種查詢方式，下面就為您介紹MySQL多表聯(lián)合查詢的語(yǔ)法，供您參考學(xué)習(xí)之用。

MySQL多表聯(lián)合查詢語(yǔ)法：

復(fù)制代碼代碼如下:

SELECT * FROM 插入表 LEFT JOIN 主表 ON t1.lvid=t2.lv_id select * from mytable,title where 表名1.name=表名2.writer ;

mysql版本大于4.0，使用UNION進(jìn)行查詢，示例如下：

SELECT `id`, `name`, `date`, '' AS `type` FROM table_A WHERE 條件語(yǔ)句……
 UNION
SELECT `id`, `name`, `date`, '未完成' AS `type` FROM table_B WHERE 條件語(yǔ)句……
 ORDER BY `id` LIMIT num;

mysql版本小于4.0，需要建立臨時(shí)表，分為三步，示例如下：

第一步：建立臨時(shí)表tmp_table_name并插入table_A中的相關(guān)記錄

復(fù)制代碼代碼如下:

$sql = "CREATE TEMPORARY TABLE tmp_table_name SELECT `id`, `name`, `date`, '完成' AS `type` FROM table_A WHERE 條件語(yǔ)句……";

第二步：從table_B中取得相關(guān)記錄插入臨時(shí)表tmp_table_name中

復(fù)制代碼代碼如下:

INSERT INTO tmp_table_name SELECT `id`, `name`, `date2` AS `date`, '未完成' AS `type` FROM table_B WHERE 條件語(yǔ)句……

第三步：從臨時(shí)表tmp_table_name中取出記錄

SELECT * FROM tmp_table_name ORDER BY id DESC

union和order by、limit區(qū)別分析

代碼示例:

CREATE TABLE `test1` (
 `id` int(10) unsigned NOT NULL AUTO_INCREMENT,
 `name` varchar(20) NOT NULL,
 `desc` varchar(100) NOT NULL,
 PRIMARY KEY (`id`)
) ENGINE=InnoDB DEFAULT CHARSET=utf8

1. 以下查詢會(huì)報(bào)錯(cuò)誤：[Err] 1221 - Incorrect usage of UNION and ORDER BY

代碼示例:

select * from test1 where name like 'A%' order by name
union
select * from test1 where name like 'B%' order by name

修改為：

代碼示例:

select * from test1 where name like 'A%'
union
select * from test1 where name like 'B%' order by name

說明，在union中，不用括號(hào)的情況下，只能用一個(gè)order by(思考：union兩邊的order by的列名不一樣時(shí)，會(huì)出現(xiàn)什么樣的結(jié)果？),這會(huì)對(duì)union后的結(jié)果集進(jìn)行排序。

修改為：

代碼示例:

(select * from test1 where name like 'A%' order by name)
union
(select * from test1 where name like 'B%' order by name)

也是可以的，這兩個(gè)order by在union前進(jìn)行。

2. 同樣

代碼示例:

select * from test1 where name like 'A%' limit 10
union
select * from test1 where name like 'B%' limit 20

相當(dāng)于：

代碼示例:

(select * from test1 where name like 'A%' limit 10)
union
(select * from test1 where name like 'B%') limit 20

即后一個(gè)limit作用于的是union后的結(jié)果集，而不是union后的select。
也可以用括號(hào)括起來(lái)，以得到預(yù)期的結(jié)果：

3. UNION和UNION ALL區(qū)別

union會(huì)過濾掉union兩邊的select結(jié)果集中的重復(fù)的行，而union all不會(huì)過濾掉重復(fù)的行。

代碼示例:

(select * from test1 where name like 'A%' limit 10)
union
(select * from test1 where name like 'B%' limit 20)

下面試一個(gè)年齡段分析的復(fù)雜sql語(yǔ)句

(
 SELECT
  '5~19' AS `age`,
  SUM(`impression`) AS impression,
  SUM(`click`) AS click,
  sum(`cost`) AS cost
 FROM
  `adgroup_age_report`
 WHERE
  (
   (
    (`age` <= 19)
    AND (`adgroup_id` = '61')
   )
   AND (`date` >= '2015-11-22')
  )
 AND (`date` <= '2017-02-20')
)
UNION
 (
  SELECT
   '20~29' AS `age`,
   SUM(`impression`) AS impression,
   SUM(`click`) AS click,
   sum(`cost`) AS cost
  FROM
   `adgroup_age_report`
  WHERE
   (
    (
     ((`age` <= 29) AND(`age` >= 20))
     AND (`adgroup_id` = '61')
    )
    AND (`date` >= '2015-11-22')
   )
  AND (`date` <= '2017-02-20')
 )
UNION
 (
  SELECT
   '30~39' AS `age`,
   SUM(`impression`) AS impression,
   SUM(`click`) AS click,
   sum(`cost`) AS cost
  FROM
   `adgroup_age_report`
  WHERE
   (
    (
     ((`age` <= 39) AND(`age` >= 30))
     AND (`adgroup_id` = '61')
    )
    AND (`date` >= '2015-11-22')
   )
  AND (`date` <= '2017-02-20')
 )
UNION
 (
  SELECT
   '40~49' AS `age`,
   SUM(`impression`) AS impression,
   SUM(`click`) AS click,
   sum(`cost`) AS cost
  FROM
   `adgroup_age_report`
  WHERE
   (
    (
     ((`age` <= 49) AND(`age` >= 40))
     AND (`adgroup_id` = '61')
    )
    AND (`date` >= '2015-11-22')
   )
  AND (`date` <= '2017-02-20')
 )
UNION
 (
  SELECT
   '50~59' AS `age`,
   SUM(`impression`) AS impression,
   SUM(`click`) AS click,
   sum(`cost`) AS cost
  FROM
   `adgroup_age_report`
  WHERE
   (
    (
     ((`age` <= 59) AND(`age` >= 50))
     AND (`adgroup_id` = '61')
    )
    AND (`date` >= '2015-11-22')
   )
  AND (`date` <= '2017-02-20')
 )