[petsc-dev] Remove calls to XXXReset() in XXXDestroy()

[Barry Smith]

On May 13, 2011, at 7:01 AM, Lisandro Dalcin wrote:

> On 13 May 2011 01:51, Barry Smith <bsmith at mcs.anl.gov> wrote:
>> 
>> On May 12, 2011, at 4:48 PM, Lisandro Dalcin wrote:
>> 
> 
> Barry, now I want to understand: Why did you coded XXXDestroy() to
> call XXXReset()? Was that just to avoid code duplication? Or is it
> there some other rationale?

   XXXReset() is a "partial" destroy. To me it makes perfect sense that the complete destroy calls the partial destroy and then frees up the rest of the stuff. It avoids code duplication or having an extra function (for example having both XXXReset() and XXXDestroy() calling XXXPartialDestroy_Private()). If we ignore your particular Python problem, what I coded was completely natural and very arguably the "right" way to code it. Having code around like


     XXXPartialDestroy_Private(XXX xxx)
       .....
       return


     XXXReset(XXX xxx)
        call XXXPartialDestroy_Private(xxx)
        return

     XXXDestroy(XXX xxx)
         call XXXPartialDestroy_Private(xxx)
         ....
         return

     without a really good reason is just plain ugly and unnecessary. Who would ever introduce that extra function? In understand from the perspective of Python why you don't like the way I did it, but I cannot understand any other reason to dislike it. 

     Barry

     I would state that any language that makes it impossible or cumbersome to have a destructor call another method on the destructed object (you claim python and C++?) is very fragile and should be rethought.

> 
> 
> -- 
> Lisandro Dalcin
> ---------------
> CIMEC (INTEC/CONICET-UNL)
> Predio CONICET-Santa Fe
> Colectora RN 168 Km 472, Paraje El Pozo
> 3000 Santa Fe, Argentina
> Tel: +54-342-4511594 (ext 1011)
> Tel/Fax: +54-342-4511169