假設我們有兩個表，需要用row_id連線，找出值相同或者不相同的資料
with temp1 as(
select 1 as row_id,1 as cola from dual
union all select 2 ,1 from dual
union all select 3 ,null from dual
union all select 4 ,null from dual
)select * from temp1;
with temp2 as(
select 1 as row_id,1 as colb from dual
union all select 2,null from dual
union all select 3,1 from dual
union all select 4,null from dual
)select * from temp2;
我們對比較結果的期望如列5，如果：
null值的存在有點繞人，
null=null 結果為null
1=null結果為null
所以必須對當欄位可為null的時候，比較需要對null進行單獨的處理
找出cola和colb相等的記錄
不使用decode，必須對null進行單獨的處理
with temp1 as(
select 1 as row_id,1 as cola from dual
union all select 2 ,1 from dual
union all select 3 ,null from dual
union all select 4 ,null from dual
),temp2 as(
select 1 as row_id,1 as colb from dual
union all select 2,null from dual
union all select 3,1 from dual
union all select 4,null from dual
)select * from temp1 a
inner join temp2 b
on a.row_id=b.row_id
where a.cola=colb
or (a.cola is null and b.colb is null) /*必須對null進行單獨處理*/
with temp1 as(
select 1 as row_id,1 as cola from dual
union all select 2 ,1 from dual
union all select 3 ,null from dual
union all select 4 ,null from dual
),temp2 as(
select 1 as row_id,1 as colb from dual
union all select 2,null from dual
union all select 3,1 from dual
union all select 4,null from dual
)select * from temp1 a
inner join temp2 b
on a.row_id=b.row_id
where decode(a.cola,b.colb,1,0)=1
找出cola和colb不相等的記錄
不使用decode，需要對null進行一大段處理
with temp1 as(
select 1 as row_id,1 as cola from dual
union all select 2 ,1 from dual
union all select 3 ,null from dual
union all select 4 ,null from dual
),temp2 as(
select 1 as row_id,1 as colb from dual
union all select 2,null from dual
union all select 3,1 from dual
union all select 4,null from dual
)select * from temp1 a
inner join temp2 b
on a.row_id=b.row_id
where a.cola!=colb
or (a.cola is null and b.colb is not null)
or (a.cola is not null and b.colb is null) /*必須對null進行單獨處理*/
with temp1 as(
select 1 as row_id,1 as cola from dual
union all select 2 ,1 from dual
union all select 3 ,null from dual
union all select 4 ,null from dual
),temp2 as(
select 1 as row_id,1 as colb from dual
union all select 2,null from dual
union all select 3,1 from dual
union all select 4,null from dual
)select * from temp1 a
inner join temp2 b
on a.row_id=b.row_id
where decode(a.cola,b.colb,1,0)=0
盡量避免在SQL語句的WHERE子句中使用函式
start 在sql語句的where子句中應該盡量避免在字段上使用函式，因為這樣做會使該字段上的索引失效，影響sql語句的效能。即使該字段上沒有索引，也應該避免在字段上使用函式。考慮下面的情況 現在要求你把2009.9.24註冊的使用者都查出來，怎麼辦？可能有人會這麼寫 不過很遺憾，這個語句是錯誤的...
sql語句中where的引號用法
因為 ip 沒有加單引號 請注意，我們在例子中的條件值周圍使用的是單引號。sql 使用單引號來環繞文字值 大部分資料庫系統也接受雙引號 如果是數值，請不要使用引號。這是正確的 select from persons wherefirstname bush 這是錯誤的 select from pers...
SQL語句中where和 on的區別
join過程可以這樣理解 首先兩個表做乙個笛卡爾積，on後面的條件是對這個笛卡爾積做乙個過濾形成一張臨時表，如果沒有where就直接返回結果，如果有where就對上一步的臨時表再進行過濾。下面看實驗 先準備兩張表 先執行inner join 結果沒有區別，前者是先求笛卡爾積然後按照on後面的條件進行...