On Saturday 26 January 2008, Vincent Torri wrote: > On Sat, 26 Jan 2008, Mike Frysinger wrote: > > On Saturday 26 January 2008, Vincent Torri wrote: > >> On Sat, 26 Jan 2008, Mike Frysinger wrote: > >>> On Saturday 26 January 2008, Vincent Torri wrote: > >>>>> #ifdef HAVE_NETINET_IN_H > >>>>> -#include <netinet/in.h> > >>>>> -#elif _WIN32 > >>>>> -#include <winsock2.h> > >>>>> +# include <netinet/in.h> > >>>>> +#endif > >>>>> +#ifdef HAVE_WINSOCK2_H > >>>>> +# include <winsock2.h> > >>>>> #endif > >>>> > >>>> it is completely useless to check it in configure script as winsowk2.h > >>>> is always avalaible on windows. So you replace one #ifdef which always > >>>> works by a check in configure + a #ifdef. Where is the gain ? > >>> > >>> so you dont have to keep adding crap for every random windows related > >>> compiler. like the random windows mobile / ce / whatever microsoft > >>> puts out. do it once, do it right, and then you're done. > >> > >> You do not it right. Before doing that, I asked mingw devs on how to do > >> that. I think that they know a bit more on how to do that than you. > >> > >>>> Please, revert what you committed. > >>> > >>> unless you can show this being incorrect, no > >> > >> With cegcc, I have to define / undefine macro (in Eet_private.h). So > >> anyway I have to check whether or not cegcc is used or not. > >> > >> And as I told you, people that know a lot more than you on windows told > >> me how I have to check things correctly. If you don't revert your > >> changes, I'll do it myself and i would like you to ask me questions and > >> waiting for my answers before commiting when dealing with windows stuff. > >> I can do mistakes, and I would be glad to know them. But here, what i > >> did here is correct. > > > > polluting multiple places with windows cruft that do the same thing over > > and over is not the way to go. if you want to maintain a mess of ifdefs, > > write it in *one* place and have all the other files pull it in. > > for example, about the mmap and fcntl calls in eet_lib.c, you would > suggest to write in eet_win32.c the ports of these functions and call them > in eet_lib.c ?
exactly ... where we can unify these pieces without touching common code the better. not sure if there's a "core" piece of e17 where we can move all of this, but if we had a few files that we could just drop into the different e libraries, that'd be a good start i think (think gnulib here). like a "system.h" where we stick all of the win32 ifdefs/includes and then everyone pulls that. your thoughts on mmap/fcntl sound like an excellent example. _e_mmap() / _e_fd_cloexec(). -mike
signature.asc
Description: This is a digitally signed message part.
------------------------------------------------------------------------- This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2008. http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
