On May 25, 2007, at 4:04 AM, Ivan Vecerina wrote: >> Personally I don't care how FLTK internally implements arrays. > I do think that using new[], delete[] and memcpy when a single > realloc would have worked is a crime ;) -- that is, one should > use a robust and maintainable C idiom, or a C++ one.
New/delete, and malloc/free/realloc must never be mixed. I assume that the code you are referring to uses new[], so using a realloc is out of the question, and there is no equivalent interface in new/delete. > There is a lot to like in FLTK, and it definitely deserves its place > as an important framework option. But some design quirks do > contribute to leaving me with a mitigated impression, and I was > hoping to see a deeper modernization effort in the 2.0 version. 2.0 is also probably around eight years old, almost as old as FLTK1. IIRC Bill started it pretty soon after publishing 1. > But how receptive would the FLTK community be to this ? You are asking at the right time. FLTK1 is going final these days and we will have some serious talks about the future of FLTK. > What is the interest in modernizing the interface ? There is big interest in that, but we have to keep in mind that a big group of users is on embedded systems. Templates, exceptions, and STL are still an issue on some of the embedded environments. > And would some internal clean-up efforts be accepted ? Certainly. Cleanup work is always welcome. Matthias ---- http://robowerk.com/ _______________________________________________ fltk mailing list [email protected] http://lists.easysw.com/mailman/listinfo/fltk
