[petsc-dev] Naming convention of examples

Barry Smith bsmith at mcs.anl.gov
Thu Feb 19 07:20:35 CST 2015

> On Feb 19, 2015, at 4:26 AM, Dave May <dave.mayhem23 at gmail.com> wrote:
> Is there a naming convention one should follow when introducing new examples?
> Specifically, in
>   src/vec/vec/examples/tutorials
> I see the following
>   ex1.c   ex10.c  ex11.c  ex12.c  ex15.c  ex16.c  ex18.c  ex19.c  ex2.c   ex21.c  ex3.c   ex42.c  ex42a.c ex5.c   ex6.c   ex7.c   ex8.c   ex9.c
> If I want to add a new example in this directory, should I 
>   (i) choose a file name based on the first availble empty slot between ex1.c and ex42.c?, or
>   (ii) choose a file name based on the largest existing test, e.g. ex43.c?; or

   Either i or ii is fine but note that the C and Fortran examples are suppose to go in pairs; so if ex1.c is not taken but ex1f.F is taken unless your new code is a C copy of ex1f.F you should skip ex1.c  In other words pick a number that is not used in either C or Fortran


>   (iii) ignore the distribution numbers of the existing examples all together and choose any number for my example which isn't already taken?
> Cheers,
>   Dave

More information about the petsc-dev mailing list