I've checked in the followinng change (among some others): configure, configure.in, include/autoconf.h, includes.h: add check for time.h and include it if we find it. I'm not sure what check is doing the more complex checks for time.h and sys/time.h compatibility. But I've added a check so that it will only include time.h if that is safe to include with sys/time.h "H. S. Teoh" wrote: > > On Tue, Mar 27, 2001 at 10:01:38PM -0800, Mark Wedel wrote: > > > > The crossfire configure scripts looks for sys/time.h, and if it finds it, > > includes it (this is done in include/includes.h) This seems to work on most all > > systems. > > > > Now if gnu/linux only has a include/time.h and not a include/sys/time.h, that > > would be the cause of the problem. It is actually fairly easy to modify the > > configure and include/includes.h file to deal with this properly. I would just > > like for this to be confirmed and not just that configure for whatever reason > > did not properly detect the sys/time.h file. > [snip] > > Actually, Linux (at least as configured on my system) has *both* > include/time.h and include/sys/time.h, and they are actually *different*. > For example, include/sys/time.h does not define struct tm (which is the > cause of the problem), but include/time.h does. > > I notice there's a flag in the code to the effect of, "safe to include > both time.h and sys/time.h" -- ./configure detects that it is safe on my > system, but for whatever reason only includes sys/time.h when actually > compiling the program. Perhaps this can be modified? > > T > > P.S. no need to Cc: me in replies, I'm on the list. > > -- > "One disk to rule them all, One disk to find them. One disk to bring them all > and in the darkness grind them. In the Land of Redmond where the shadows > lie." -- The Silicon Valley Tarot > _______________________________________________ > crossfire-devel mailing list > crossfire-devel at lists.real-time.com > https://mailman.real-time.com/mailman/listinfo/crossfire-devel