[petsc-dev] printing matlab vectors

[Mark Adams]

Oh, duh, I put the name on the viewer and not the vector.

On Fri, Jun 12, 2015 at 8:54 AM, Mark Adams <mfadams at lbl.gov> wrote:

> I do this:
>
>   call
> VecCreateSeqWithArray(PETSC_COMM_SELF,ione,n64,phi,vec,ierr);CHKERRQ(ierr)
>   call PetscViewerASCIIOpen(PETSC_COMM_SELF, file_name,
> viewer,ierr);CHKERRQ(ierr)
>   call
> PetscViewerSetFormat(viewer,PETSC_VIEWER_ASCII_MATLAB,ierr);CHKERRQ(ierr)
>   call PetscObjectSetName(viewer,vec_name,ierr);CHKERRQ(ierr)
>   call VecView(vec,viewer,ierr);CHKERRQ(ierr)
>
> But I still get this:
>
> 05:51 nid02816 new_em *= ~/epsi/XGC-1_examples/em_tearing$ head xgc_lhs.m
> %Vec Object: 1 MPI processes
> %  type: seq
> Vec_0xc4000009_0 = [
> -0.0000000000000000e+00
> 5.6929682630215887e-16
>
> Am I missing something?  I want Vec_0xc4000009_0 to be 'vec_name'
>
> Mark
>
>
> On Thu, Jun 11, 2015 at 8:06 PM, Jed Brown <jed at jedbrown.org> wrote:
>
>> Mark Adams <mfadams at lbl.gov> writes:
>>
>> > I am printing matlab vectors and I have a matlab script to read these
>> > files, take differences, etc.  PETSc name the vectors
>> > (eg, Vec_0xc4000009_0) in a non-deterministic way as far as I can
>> > tell.
>>
>> PetscObjectSetName()
>>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20150612/a11482b5/attachment.html>