You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user-java@ibatis.apache.org by Guy Rouillier <gu...@burntmail.com> on 2009/10/05 06:14:41 UTC
Select one set null instead of throwing exception
Currently, performing a single row select throws an exception if zero or
> 1 rows are returned. While this is certainly workable, an
alternative that would be easier on the programmer would be to set the
result to null if no rows are found, and only throw an exception if > 1
rows are found.
The documentation suggests doing a count to determine what situation you
are in, then issuing the appropriate select once you know. But that
will result in an extra trip to the DB. In a common scenario where you
are querying a primary key, you know that you will get zero or one row
back. So rather than having to put the single row select into a
try-catch block, a more natural paradigm would be to assign the result
and then check the result for null.
--
Guy Rouillier
---------------------------------------------------------------------
To unsubscribe, e-mail: user-java-unsubscribe@ibatis.apache.org
For additional commands, e-mail: user-java-help@ibatis.apache.org
Re: Select one set null instead of throwing exception
Posted by Clinton Begin <cl...@gmail.com>.
We've removed that exception anyway. :-)
Popular community vote.
On Mon, Oct 5, 2009 at 12:07 AM, Guy Rouillier <gu...@burntmail.com>wrote:
> Ignore this. I realized after taking a break that it doesn't work with
> primitive types. I think the thing I find most frustrating with Java is
> that it doesn't treat primitive types in a consistent manner with classes
> derived from Object.
>
>
> Guy Rouillier wrote:
>
>> Currently, performing a single row select throws an exception if zero or
>> > 1 rows are returned. While this is certainly workable, an alternative
>> that would be easier on the programmer would be to set the result to null if
>> no rows are found, and only throw an exception if > 1 rows are found.
>>
>> The documentation suggests doing a count to determine what situation you
>> are in, then issuing the appropriate select once you know. But that will
>> result in an extra trip to the DB. In a common scenario where you are
>> querying a primary key, you know that you will get zero or one row back. So
>> rather than having to put the single row select into a try-catch block, a
>> more natural paradigm would be to assign the result and then check the
>> result for null.
>>
>>
>
> --
> Guy Rouillier
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: user-java-unsubscribe@ibatis.apache.org
> For additional commands, e-mail: user-java-help@ibatis.apache.org
>
>
Re: Select one set null instead of throwing exception
Posted by Guy Rouillier <gu...@burntmail.com>.
Ignore this. I realized after taking a break that it doesn't work with
primitive types. I think the thing I find most frustrating with Java is
that it doesn't treat primitive types in a consistent manner with
classes derived from Object.
Guy Rouillier wrote:
> Currently, performing a single row select throws an exception if zero or
> > 1 rows are returned. While this is certainly workable, an
> alternative that would be easier on the programmer would be to set the
> result to null if no rows are found, and only throw an exception if > 1
> rows are found.
>
> The documentation suggests doing a count to determine what situation you
> are in, then issuing the appropriate select once you know. But that
> will result in an extra trip to the DB. In a common scenario where you
> are querying a primary key, you know that you will get zero or one row
> back. So rather than having to put the single row select into a
> try-catch block, a more natural paradigm would be to assign the result
> and then check the result for null.
>
--
Guy Rouillier
---------------------------------------------------------------------
To unsubscribe, e-mail: user-java-unsubscribe@ibatis.apache.org
For additional commands, e-mail: user-java-help@ibatis.apache.org