On 20 Feb 2013, at 18:50, Michael Baeuerle wrote: > Greg Ercolano wrote: >> On 02/20/13 05:15, Michael Baeuerle wrote: >>> >>> Looks like all #ifdef cases are verified now except the ones for Digital >>> and SGI Unix. Unfortunately I don't have access to such machines. >> >> I'll check SGI (IRIX 6.5) and follow up in the STR. > > Albrecht has mentioned the ABI and it looks like this is a problem for > 'fl_filename_list()'. The sort functions are exported as C, but > 'fl_filename_list()' is exported as C++. > I have tried to statically link code that use the new 'filename.H' > against an older library and this fails because the parameter types of > the C++ function do not exactly match. For the C functions it works and > there is not even a warning (as expected because I have restricted what > the functions are allowed to do with the parameters). > > I'm not a C++ expert, is it possible to overload the function > declaration without breaking other things?
I suspect the "best" option here is possibly to revert the function prototype to the non-const version...? I mean, change the typedef for Fl_File_Sort_F back to being non-const... Would that do the trick, I wonder. It'd be less const-correct than the proposed code, but no worse than what we had before. Thoughts? _______________________________________________ fltk mailing list [email protected] http://lists.easysw.com/mailman/listinfo/fltk
