ibatis-user-java mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Michael Zurke <mis...@zurke.info>
Subject Re: canceling statement?
Date Mon, 18 Apr 2005 15:11:14 GMT
as far as i know the only way to do it on the jdbc layer. the Statement 
interface supplies a cancel() method for that purpose. the 
database/driver has supprt that functionality, i guess.

mish

Sasha Borodin wrote:
> Is there an sql command you can issue to cancel an already running sql
> statement?  Is there an abstract way to identify a statement running in a db
> system, which could be gotten through iBatis?
> 
> -Sasha
> 
> 
> 
>>From: Michael Zurke <mischa@zurke.info>
>>Reply-To: <ibatis-user-java@incubator.apache.org>
>>Date: Mon, 18 Apr 2005 15:57:39 +0100
>>To: <ibatis-user-java@incubator.apache.org>
>>Subject: Re: canceling statement?
>>
>>yes - i think your're right when it comes to the multiple query/combine
>>result issue - using DAO in this case would definitely make more sense.
>>but in case DAO doesn't fit my model i still want to be able to built my
>>own 'multilpe query executor' so i would be nice if i could cancel
>>statments on the SQLMap layer. but maybe the problem is too specific to
>>support it on a generic framework level. anyway ;)
>>
>>Brice Ruth wrote:
>>
>>>I could see this being done in a DAO layer, perhaps ... I'm not sure
>>>that the SqlMap layer is necessarily the right place for it. What do you
>>>think?
>>>
>>>On 4/18/05, *Michael Zurke* <mischa@zurke.info
>>><mailto:mischa@zurke.info>> wrote:
>>>
>>>    i think it would make sense if ibatis would support something like
>>>    multiple query processing (e.g. a query executed in parallel on
>>>    different database clusters. then the abstraction layer should combine
>>>    the results. in that case it would be handy to specify a timeout to get
>>>    results back from the most responsive clusters only (-> cancel). but
>>>    this is theoretical since ibatis doesn't support such functionality
>>>    (right?). but still - if there would be something like SQLMap.cancel()
>>>    one could implement the parallel querying quite easy, i guess.
>>>
>>>    mish
>>>
>>>    Brandon Goodin wrote:
>>>
>>>>It is not a funciton that is available. But, i'm not sure why it
>>>>wouldn't be available. iBatis uses straight JDBC underneath. It could
>>>>provide some functionlaity to cancel. But, i'm not sure how we would
>>>>provide the criteria for a cancel. Please expound on "why" you would
>>>>cancel.
>>>>
>>>>Brandon
>>>>
>>>>On 4/18/05, Brice Ruth <bdruth@gmail.com
>>>
>>>    <mailto:bdruth@gmail.com>> wrote:
>>>
>>>>>iBATIS is thread-local. The cancel() method requires that a
>>>
>>>    separate thread
>>>
>>>>>cancel a statement in an executing thread. I don't believe this
>>>
>>>    would be
>>>
>>>>>supported in this type of framework.
>>>>>
>>>>>Reference:
>>>>>http://java.sun.com/j2se/1.4.2/docs/api/java/sql/Statement.html#cancel()
>>>>>
>>>>>BR
>>>>>
>>>>>
>>>>>
>>>>>On 4/18/05, Michael Zurke <mischa@zurke.info
>>>
>>>    <mailto:mischa@zurke.info>> wrote:
>>>
>>>>>>is it somehow possible to cancel an executed query via SQLMaps?
>>>>>>as far as i know PreparedStatement has a cancel() method for
>>>
>>>    that puropse.
>>>
>>>>>
>>>>>
>>>>>--
>>>>>Brice Ruth
>>>>>Software Engineer, Madison WI
>>>>
>>>>
>>>
>>>
>>>
>>>-- 
>>>Brice Ruth
>>>Software Engineer, Madison WI
> 
> 
> 
> 


Mime
View raw message