sql problem more difficult to appear in the forum: 46 (date conditions appear strange questions)

Original: SQL problem more difficult to appear in the forum: 46 (date conditions appear strange questions)

Recently, in the forum, I met a lot more difficult sql problem, although they can be resolved, but found a few days later, they can not remember, forget the solution of.

So, I feel the need to be recorded, so that after the encounter this problem again, and get answers from the idea.

Strange issue date conditions arise.

http://bbs.csdn.net/topics/390498925 

select top 1 ekeyid,ProbeMaterial,Result,LabTestDate from PatientLabTestResults
where ProbeMaterial='Ca'  
  and LabTestDate between convert(datetime,2013/1/1) and
   convert(datetime,'2013/6/24') order by LabTestDate desc

这样查的出结果，变成
select top 1 ekeyid,ProbeMaterial,Result,LabTestDate from PatientLabTestResults 
where ProbeMaterial='Ca'  
  and LabTestDate between convert(datetime,2013/1/1) and
   convert(datetime,2013/6/24) order by LabTestDate desc

So after not find the results, in fact, just removed 2013/6/24's' number only, that we can understand what is the problem?

   
   
if OBJECT_ID('t') is not null
drop table t
go

create table t(d datetime)


insert into t
select cast( '2013-05-01' as datetime) as d
union all
select cast( '2013-05-10' as datetime)


/*
1.

通过查询计划能看出，SQL Server把下面的查询转化成了：

select * from t where d >= convert(datetime,2013/5/1,0) and d <= convert(datetime,'2013/6/24',0)

也就是查询条件：
d >= convert(datetime,2013/5/1,0) and d <= convert(datetime,'2013/6/24',0)

进一步转化：
d >= '1901-02-07 00:00:00.000' and d <= 2013-06-24 00:00:00.000

这样就能查询出结果集。
*/
select *
from t
where d between convert(datetime, 2013/ 5/ 1) and convert(datetime, '2013/6/24')



/*
2.

通过查询计划能看出，SQL Server把下面的查询转化成了：

select * from t where d >= convert(datetime,2013/5/1,0) and d <= convert(datetime,2013/6/24,0)

也就是查询条件：
d >= convert(datetime,2013/5/1,0) and d <= convert(datetime,2013/6/24,0)

进一步转化：
d >= '1901-02-07 00:00:00.000' and d <= '1900-01-14 00:00:00.000'

由于SQL Server 把2013/6/24中的斜杠，当成了除号，也就是按除法计算了，
比如：2013/6/24 就等于13，那么由于datetime默认值是默认值： 1900-01-01 00:00:00，
那么加上13后，就是1900-01-14 00:00:00.000，这样后就查不出结果了.
*/
select *
from t
where d between convert(datetime, 2013/ 5/ 1) and convert(datetime, 2013/ 6/ 24)
  
  

In fact,

2013/5/1 is a division operation, the result is 402.
After 2013/6/24 do the divisions, it is 13.

Since the default data type is datetime: "1900-01-01 00:00:00",

Therefore, the above convert (datetime, 2013/5/ 1) is' 1900-01-01 00:00:00 'plus 402,
that is,' 1901-02-0700: 00: 00.000 ",

And convert (datetime, 2013/6/ 24) that is' 1900-01-01 00:00:00 'plus 13,
that is,' 1900-01-1400: 00: 00.000 ",

So you will not find the results.

Therefore, the above should be rewritten 2013/5/1 "2013/5/1", must be in quotation marks.