> On 18 Jan 2019, at 18:43, Jed Brown <j...@jedbrown.org> wrote: > > "Smith, Barry F. via petsc-dev" <petsc-dev@mcs.anl.gov> writes: > >> I'm very hesitant to have the -viewer_format supported by PETSc >> because it reintroduces the concept of "setting" the viewer format, >> as opposed to pushing and popping formats. I think it is a mistake >> to have both the ability to set the format and to push/pop it, they >> don't play well together. > > I think we could allow setting the format *before* any push takes place. > I.e., if a caller creates a Viewer, they could set its format at that > moment, but anyone who is given a Viewer to use can only push/pop. > > I don't know that such behavior is necessary/desirable in this instance.
For me it would make sense. It would allow to set the default format from options while not spoiling the push/pop. But what do you Barry think? Or alternatively, what about PetscViewerPushFormatFromOptions(PetscViewer) I proposed earlier? Thanks Vaclav
