On Tue, Oct 28, 2008 at 05:14:00PM +0100, [EMAIL PROTECTED] wrote: > >It seems bogus, but if I recall correctly, the standards conformance > >gurus say that we're not allowed to walk into the user's name space in > >this way: those symbols are defined by the standards, so they must not > >exist. If you don't like that, then don't compile in a standards- > >conformant environment. > > But if you add -D__EXTENSIONS__, you're free from standards conformance, > right?
Yes, but uttering "-D_XOPEN_SOURCE=500 -D__EXTENSIONS__" is oxymoronic, since you're saying you both want a particular standards environment, and you don't. These defines don't mean "turn on standards behaviour", they mean "I am an application that complies with this standard". This is a generic problem with our headers, where useful implementations are hidden behind the standards guards. The _B_FALSE thing is really just a symptom. The better solution here is to make the socket structure also dependent on some other define and use that. I suggested in IRC some kind of I_WANT_LATEST define that would always be current (and thus could always break your legacy code) to solve this kind of issue. Fixed the _B_FALSE thing is just papering over the issue IMHO. regards john _______________________________________________ opensolaris-code mailing list opensolaris-code@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/opensolaris-code
