On Jan 25 12:16, Eli Zaretskii wrote: > > From: Angelo Graziosi <[EMAIL PROTECTED]> > > > > With Cygwin DLL 1.5.25 (and the changes to 'tz...', I suppose), building > > Emacs from CVS gives this warning: > > > > ... > > gcc -c -Demacs -DHAVE_CONFIG_H -DUSE_LUCID -I. -I/tmp/emacs/src > > -I/usr/X11R6/include -g -O2 -Wno-pointer-sign /tmp/emacs/src/strftime.c > > > > /tmp/emacs/src/strftime.c:58: warning: '_tzname' redeclared without > > dllimport attribute: previous dllimport ignored > > ...
Btw., nothing has changed in the way tzname (_tzname) is declared in newlib's time.h header for years. > > It seems that the following simple patch avoid it: > > ================================================= > > --- strftime.c.orig 2007-01-14 04:24:37.000000000 +0100 > > +++ strftime.c 2008-01-24 12:04:40.031250000 +0100 > > @@ -55,9 +55,13 @@ > > #endif > > #if HAVE_TZNAME > > #ifndef USE_CRT_DLL > > +#ifdef __CYGWIN__ > > +extern __IMPORT char *tzname[]; > > +#else > > extern char *tzname[]; > > #endif > > #endif > > +#endif > > Thanks. But it's better to put this kind of platform-dependent stuff > in src/s/cygwin.h, not in general source files. Would that work for > you? As I wrote in my reply on the Cygwin list, there should be no reason to declare tzname on POSIX conformant systems at all. Just including <time.h> is the way to go. The extern declaration should be guarded by an OS-agnostic configure test which checks whether tzname is declared in time.h or not. Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Project Co-Leader cygwin AT cygwin DOT com Red Hat -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
