about viewer formats

[Barry Smith]

On Nov 3, 2008, at 1:36 PM, Lisandro Dalcin wrote:

> On Mon, Nov 3, 2008 at 2:34 PM, Barry Smith <bsmith at mcs.anl.gov>  
> wrote:
>>
>>  Done and pushed.
>
> OK. Many thanks,
>
>>
>> I guess at one time we had it in our heads that NO format names  
>> would be
>> shared among different
>> viewers, but now I don't see why some basic ones cannot be shared.
>
> BTW, Does it make sense to require the type_name being already set (I
> mean, using the PetscValidType macro) before setting the format?

   Hmm, it will likely get overwritten by the PetscViewerCreate_XXX()  
and
hence be ignored and thus confusing to users. Thus for user friendlyness
maybe we should not allow early setting.

    Looking at a few of the PetscViewerCreate_XXX() I see that we are  
inconsistent
in setting the viewer->format or not. This is bad, we should either  
always set
it, or never set it (and only have it set in the  
PetscViewerCreate()).????
Will you change the code to always set it in each  
PetscViewerCreate_XXX()
please?


   Barry



>
>
>> On Nov 3, 2008, at 7:50 AM, Lisandro Dalcin wrote:
>>
>>> Barry, I saw you commit removing PETSC_VIEWER_BINARY_NATIVE. This
>>> makes me think that PETSC_VIEWER_ASCII_DEFAULT and
>>> PETSC_VIEWER_BINARY_DEFAULT (which is NOT used in sources) are a bit
>>> redundat, Why not to remove both and add a new PETSC_VIEWER_DEFAULT?
>>> Does it make sense?
>>
>>
>
>
>
> -- 
> Lisandro Dalcín
> ---------------
> Centro Internacional de Métodos Computacionales en Ingeniería (CIMEC)
> Instituto de Desarrollo Tecnológico para la Industria Química (INTEC)
> Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET)
> PTLC - Güemes 3450, (3000) Santa Fe, Argentina
> Tel/Fax: +54-(0)342-451.1594
>