From mint-bounce@lists.fishpool.fi Thu Feb 19 06:55:47 2009 Date: Thu, 19 Feb 2009 11:52:50 +0000 (GMT) From: David Brownlee To: Alan Hourihane cc: Frank Naumann , mint@fishpool.com Subject: Re: [MiNT] timegm() on MiNT In-Reply-To: <1235037052.24227.35.camel@jetpack.demon.co.uk> Message-ID: References: <1235037052.24227.35.camel@jetpack.demon.co.uk> User-Agent: Alpine 2.00 (NEB 1167 2008-08-23) MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed X-ecartis-version: Ecartis v1.0.0 Sender: mint-bounce@lists.fishpool.fi Errors-to: mint-bounce@lists.fishpool.fi X-original-sender: abs@absd.org Precedence: bulk List-help: List-unsubscribe: List-Id: X-List-ID: List-subscribe: List-owner: List-post: On Thu, 19 Feb 2009, Alan Hourihane wrote: > On Thu, 2009-02-19 at 08:48 +0100, Frank Naumann wrote: >> Hello! >> >>> /usr/include/time.h provides a prototype for timegm() but >>> it does not exist in libc. This breaks the configure script >>> for bmake, because it finds timegm() missing, so defines its >>> own, and the prototype it generates conflicts with the >>> one in /usr/include/time.h. >>> >>> Am I missing something - is there a timegm() in another library? >>> If not, would it be possible to comment out the definitions in >>> time.h? >> >> Looks like a failure in the header files as timegm() is not implemented >> inside the MiNTLib. >> >> So we need to implement it or remove the definition from time.h :-) > > This signals a larger problem. > > mktime() is slightly broken in our current implementation. And timegm.c > does exist in time/ but it isn't compiled as it references glibc's way > of doing things. > > We should be able to yank in the glibc equivalents of these and then > remove the timezone files and use the external packages. Could the definitions just be put under #ifndef __MINT__ for now so it doesn't confuse applications?