[MOAB-dev] Creation of parallel tags

Jed Brown jed at 59A2.org
Mon Mar 2 10:29:19 CST 2009


On Mon, Mar 2, 2009 at 16:19, Jason Kraftcheck <kraftche at cae.wisc.edu> wrote:

> A few comments about your patches.
>
> mhdf.h:
>
> The change from C++ to C-style comments shouldn't be necessary within the
> '#ifdef __cplusplus' blocks.  But it obviously doesn't hurt anything either.
>  Are there some compiler warnings or some such thing that motivated this change?

Yes, gcc-4.3.3 was giving warnings when compiling ANSI C source.  (I
know it is protected, but maybe the parser is removing comments before
the prepocessor?)

> iMesh_MOAB.cpp:
>
> The change to the allowable length of the returned string isn't quite the
> right way to fix that bug.  The ITAPS interfaces to not guarantee
> null-terminated strings.  It is the "descr[len] = '\0';" statement that is
> incorrect and should be removed.

I didn't know that about returning non-terminated strings.  I agree
that there are better ways to fix it anyway, but what was there could
obviously write beyond the array.

> Why to you change the default storage order?

Because interleaved is the native storage order. (I don't query the
interface that way, but I believe my change is what was intended.)

Jed


More information about the moab-dev mailing list