MySQL拆分字符串函数SPLIT_STR不适用于LEFT JOIN。有问题吗?

时间:2012-01-02 20:58:51

标签: mysql sql join user-defined-functions

创建MySQL拆分字符串函数SPLIT_STR fedecarg.com/.../mysql-split-string-function/

CREATE FUNCTION SPLIT_STR(
  x VARCHAR(255),
  delim VARCHAR(12),
  pos INT
)
RETURNS VARCHAR(255)
RETURN REPLACE(SUBSTRING(SUBSTRING_INDEX(x, delim, pos),
       LENGTH(SUBSTRING_INDEX(x, delim, pos -1)) + 1),
       delim, '');

运行SQL:

SELECT t.en AS `type`, SPLIT_STR(l.en, ',', 1) AS city,
SPLIT_STR(l.en, ',', 2) AS country
FROM table1
JOIN table2
USING ( id )
LEFT JOIN table3 AS t ON table1.type = t.id
/* the next line has failure with SPLIT_STR */
LEFT JOIN table3 AS l ON table1.location = l.id
WHERE language.lang =  'en'
AND language.url =  'europe-countries'
LIMIT 1;

表1

id               | type            | location
-----------------+-----------------+-----------------
6BC45C02         | place           | london,england

表2

id               | url
-----------------+-----------------
6BC45C02         | europe-countries

表3

id               | en
-----------------+-----------------
london           | London
england          | England

结果失败:

type             | city            | country
-----------------+-----------------+----------------
place            | NULL            | NULL

预期结果将是返回citycountry

type             | city            | country
-----------------+-----------------+-----------------
place            | London          | England

检查SPLIT_STR是否使用简单的SQL:

SELECT SPLIT_STR(location, ',', 1) AS city, SPLIT_STR(location, ',', 2) AS contry
FROM table1
WHERE id =  '6BC45C02'
LIMIT 1;

它返回了很好的结果:

city             | contry
-----------------+-----------------
london           | england

2 个答案:

答案 0 :(得分:1)

也许这......但表现会很糟糕。

SELECT T1.type,
  SPLIT_STR(t.en, ',', 1) AS city,
  SPLIT_STR(l.en, ',', 2) AS country
FROM table1 t1
INNER JOIN table2 t2 
  ON T1.ID = T2.ID
LEFT JOIN table3 t 
  ON t.id = SPLIT_STR(t1.location, ',', 1)
LEFT JOIN table3 l 
  ON l.id = SPLIT_STR(t1.location, ',', 2)
WHERE t2.url = 'europe-countries'
LIMIT 1;

这样会更好,因为它不需要调用函数4次:(使用coalesce来确定函数是否按预期工作将返回正确的情况,然后是小写,然后如果函数没有按预期工作则函数会中断)

SELECT InTable.Type, 
   coalesce(t.en, inTable.City, 'FunctionBroke') as city, 
   coalesce(l.en, intable.country, 'FunctionBroke2') as country
FROM 
    (SELECT t1.type,
      SPLIT_STR(T1.Location, ',', 1) AS City,
      SPLIT_STR(T1.Location, ',', 2) AS Country
    FROM table1 T1
    INNER JOIN table2 T2 
      ON T1.ID = T2.ID 
        AND t2.url='europe-countries'
    ) InTable
LEFT JOIN table3 t 
  ON InTable.City = t.id
LEFT join table3 l 
  ON InTable.Country = l.id
LIMIT 1;

更好的存在:表3的唯一目的是正确判断城市/国家名称;并且是UDF(用户定义函数)分割值的唯一目的吗?

  • 编辑1首先固定的应该是t.en而不是左边的l.en 加入。
  • 编辑2固定第二应该是t1(复制和粘贴为 你)
  • 编辑3对两个回复中的所有内容感到欣慰,复制了OP中的几个错误

答案 1 :(得分:1)

从我看到的情况来看,SPLIT_STR(l.en, ',', 1)始终为null(在table3.en中没有任何内容可以拆分)另外,表table1.location = l.id的连接条件对于您的数据始终为false (londonengland都不等于london,england)。

根据您发布的所需输出,我认为您正在寻找类似的东西(我不确定什么是language.url = 'europe-countries',我在问题中没有看到名为“language”的表或别名,所以我只是忽略了它)

SELECT t1.`type` AS `type`, 
MAX(CASE WHEN t3.id = SPLIT_STR(t1.location, ',', 1) THEN t3.en END) as `city`,
MAX(CASE WHEN t3.id = SPLIT_STR(t1.location, ',', 2) THEN t3.en END) as `country`
FROM table1 t1
INNER JOIN table2 t2 ON (t2.id = t1.id)
LEFT JOIN table3 t3 ON 
 (t3.id = SPLIT_STR(t1.location, ',', 1) OR t3.id = SPLIT_STR(t1.location, ',', 2))
GROUP BY t1.`type`

已更新(l.en已被t1.location替换)