Regarding missing trunc(): I would prefer a configure + CMake test whether trunc() is available rather than system macros to decide if we need a replacement function. It may turn out that other systems lack this function as well and a configure/CMake test would catch all of them.

I apologize because I'm a bit late with this aspect, however this is not only true for this patch but for all other missing functions we have encountered so far or may find in the future. Looking at this particular patch made me think: "this is too complicated, it can't be right!".

I admit that I accepted platform tests like #ifdef __hpux before (and in some cases this may be OK) but we should "think" more in "feature tests" rather than "guessing" (depending on platform macros).

—
Reply to this email directly, view it on GitHub, or unsubscribe.
You are receiving this because you are subscribed to this thread.Message ID: <fltk/fltk/issues/944/2018774174@github.com>