#197: sfd support for r.terraflow --------------------------+------------------------------------------------- Reporter: adanner | Owner: [email protected] Type: defect | Status: new Priority: major | Milestone: 6.4.0 Component: Raster | Version: svn-develbranch6 Resolution: | Keywords: terraflow r.terraflow Platform: Unspecified | Cpu: Unspecified --------------------------+------------------------------------------------- Comment (by glynn):
Replying to [comment:3 adanner]: > > However: is it really necessary for the new file to be in C instead of C++? It would simplify matters to keep everything in the same language. > > Yes, the new file must be in C and not C++. The g++-4.2 throws a warning if you try to assign a string constant to a char *. E.g., Sorry; I didn't notice that you had already mentioned that in the original report. Personally, I wouldn't be especially concerned about the warnings. Especially as they are quite legitimate; those fields should really be "const char *", as they normally get assigned from string literals. Although string literals are treated as "char *", on any platform we care about, you will get a segfault if you try to modify them. The only fields which are likely to be modified are answer and answers, and even that's dubious in terms of clarity. Code should ideally copy the values to a separate buffer rather than modifying them in-place. Moving the code from C++ to C is simply hiding a legitimate issue. The compiled code will be essentially the same regardless of language, and the problems are the same. It's just that, for historical reasons, C doesn't complain while C++ does. My attitude to warnings is to either fix them for real or leave them there, with "hiding" them being the worst choice. -- Ticket URL: <http://trac.osgeo.org/grass/ticket/197#comment:4> GRASS GIS <http://grass.osgeo.org>
_______________________________________________ grass-dev mailing list [email protected] http://lists.osgeo.org/mailman/listinfo/grass-dev
