[petsc-dev] PETSC_TRUE and PETSC_TRUTH

[Chetan Jhurani]

The conversation went in direction of bool and incompatibility.  Not that there
is anything wrong with it.

My suggestion was to change only PETSC_TRUTH to something like PETSC_TRUTH_DATA
or PETSC_TRUTH_TYPE which would not be a very inconsistent change in terms of
nouns and it will avoid bugs due to typos.

Unlike PETSC_TRUE, PETSC_TRUTH looks like an implementation detail.  It is used
in only three C files in the sys directory.

Chetan

> -----Original Message-----
> From: petsc-dev-bounces at mcs.anl.gov [mailto:petsc-dev-bounces at mcs.anl.gov] On Behalf Of Lisandro
> Dalcin
> Sent: Friday, September 03, 2010 7:58 AM
> To: For users of the development version of PETSc
> Cc: petsc-dev
> Subject: Re: [petsc-dev] PETSC_TRUE and PETSC_TRUTH
> 
> On 2 September 2010 20:02, Jed Brown <jed at 59a2.org> wrote:
> > On Thu, 2 Sep 2010 18:08:12 -0400, Kai Germaschewski <kai.germaschewski at unh.edu> wrote:
> >> I know that bool is not standard (in plain old C). However, as you say, it
> >> is in C++ and C99, and in as extension in some older compilers. I have no
> >> doubt there was a good reason to introduce the type at the time, but that
> >> doesn't necessarily mean you have to keep carrying it along forever, in
> >> particular if you're changing the wording anyway. The one issue I do see is
> >> if an existing application has, e.g., typedef int bool; while the petsc
> >> header says typedef enum { false, true } bool;
> >
> > FWIW, neither of these are valid definitions, they have a different size
> > than the bool in stdbool.h (technically the size of the enum is
> > implementation-dependent, but it is almost certainly the same size as
> > int, while bool is 1 byte).
> >
> > Jed
> >
> 
> 
> 1) Is bool 1 byte in OS X?
> 
> 2) Microsoft compilers do not support C99, then there is no bool
> AFAIK, it would need emulation (with a char?)
> 
> 3) How a change would impact Fortran?
> 
> 4) What about cross-platform I/O compatibility?
> 
> 5) How to communicate them with MPI(-1) in possibly heterogeneous environments?
> 
> For all these reasons and others I'm likely missing, I think
> PetscTruth should be an enum (it will likely match an int in almost
> all platforms, even in C++), and perhaps It would be better to call it
> PetscBool.  Other way would be typedef'inf it as a 'unsigned char' and
> #define TRUE and FALSE, but not sure how to handle Fortran. Using a
> (unsigned) char would be nice because of sizeof 1, and there are no
> endian issues, thus simplifying I/O.
> 
> 
> --
> Lisandro Dalcin
> ---------------
> CIMEC (INTEC/CONICET-UNL)
> Predio CONICET-Santa Fe
> Colectora RN 168 Km 472, Paraje El Pozo
> Tel: +54-342-4511594 (ext 1011)
> Tel/Fax: +54-342-4511169