Re: RE : [Springframework-developer] Problem with MySqlMaxValueIncrementer

[Isabelle M. <isa...@me...>]

Hi JP,

As a matter of fact I also thought of requiring the equivalent of EJB's Requires New transaction. The problem still is how to implement this: how do I know whether the code is running in a CMP/BMP EJB, under JTA, ...
I'm a bit baffled, don't know the answer.

There is a solution which makes the whole implementation redundant : for the user to cast the Statement or PreparedStatement to a com.mysql.Statement (PreparedStatement) and then call getLastInsertId(). I don't want to do this in the framework, because then compilation becomes dependent on the mysql JDBC library.

Isabelle

On Fri, May 30, 2003 at 04:29:30PM +0200, JP Pawlak wrote:
> Hi Isabelle,
> 
> Ah, the continuously problematic MaxValueIncrementer ;-) !
> As with the key search we will always have potentially two Sql orders,
> how manage this?
> The best would be to have the possibility to treat separately. It would
> be convenient to assure the key search being out of any transaction and
> being committed regardless of a current transaction. In this case,
> simply the key(s) used by the rollbacked operations are lost. I don't
> know, so quickly, if it is possible and how.
> 
> Jean-Pierre
>  
> 
> > -----Message d'origine-----
> > De : spr...@li... 
> > [mailto:spr...@li...]
> >  De la part de Isabelle Muszynski
> > Envoyé : vendredi 30 mai 2003 15:28
> > À : spr...@li...
> > Objet : [Springframework-developer] Problem with 
> > MySqlMaxValueIncrementer
> > 
> > 
> > Hi everyone,
> > 
> > There is a problem with MySQLMaxValueIncrementer (and I think 
> > with the Oracle one as well). Currently the code doesn't do 
> > any transaction management which has 3 consequences :
> > (1) if the calling code does a rollback, the value(s) cached 
> > in / handed out by the object are invalid
> > (2) the code requires the caller to do a commit, which I'm 
> > sure most people will forget to do if using the framework.
> > (3) when requesting a bunch of keys, it is possible to get a 
> > series of non-sequential numbers from the database. This 
> > because of the loop in the code, which is requred by mySQL 
> > last_insert_id() semantics. This particular problem could be 
> > solved by maintaining an array of values, instead of just a 
> > maxId. I'll make this change over the weekend. This 
> > particular problem won't show up in the Oracle 
> > implementation. I'm not sure how to solve the items 1 & 2 
> > though. The code can be called in a variety of circumstances 
> > (BMP and CMP beans, no EJB, ...). Can we even assume the 
> > presence of a transaction manager?
> > 
> > Any ideas?
> > 
> > 
> > -- 
> > Isabelle Muszynski
> > Software Engineer
> > Zandweellaan 4
> > 2660 Antwerpen
> > Belgium
> > Tel. 32-(0)3-830 18 54
> > Mobile: 32-(0)485 49 50 89
> > Email: isa...@me...
> > Website: www.meta-logix.com
> > 
> > 
> > -------------------------------------------------------
> > This SF.net email is sponsored by: eBay
> > Get office equipment for less on eBay! 
> > http://adfarm.mediaplex.com/ad/ck/711-11697-> 6916-5
> > 
> > _______________________________________________
> > 
> > Springframework-developer mailing list 
> > Spr...@li...
> > https://lists.sourceforge.net/lists/listinfo/springframework-developer
> > 
> 
> 
> 
> 

-- 
Isabelle Muszynski
Software Engineer
Zandweellaan 4
2660 Antwerpen
Belgium
Tel. 32-(0)3-830 18 54
Mobile: 32-(0)485 49 50 89
Email: isa...@me...
Website: www.meta-logix.com