On Fri, Sep 28, 2012 at 6:16 PM, Karl Rupp <span dir="ltr"><<a href="mailto:rupp@mcs.anl.gov" target="_blank">rupp@mcs.anl.gov</a>></span> wrote:<br><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
b) typedef const char* VecType; imposes usability restrictions and would hence break a lot of user-code.</blockquote></div><br><div>The thing is, the public interface always uses const VecType. The place this would be most likely to break user code is if they had</div>
<div><br></div><div>const VecType vtype;</div><div>...</div><div>vtype = "mpi";</div><div><br></div><div><br></div><div>I am in favor of an eventual rename of everything, but I always hated the #define VecType char* business. Our implementation and private functions that manipulate a type can use raw char* if they need a mutable "VecType".</div>