> Chuck Kirschman wrote:
> >
> > This patch seems to fix the issue with the time change in CVS on NT.
> > Currently, every file is considered changed and sent up to the server for
> > comparison. This is only noticeable by the time required and certain
>binary
> > files which, due to a different bug, are read as ascii and considered to
>be
> > of a different size.
>
> I have done a bit of testing, and with a bit of help from Alexandre
>Parenteau have
> reached the conclusion that the gmtime() function is behaving correctly.
>However
> the C library is defaulted to pacific standard time unless you explicitly
>change it
> (this
> would affect localtime() output amongst other things). The fix is to put a
>call to
> _tzset() during
> initialisation on Windows machines.
>
Hey Tony. I know this is an old email (April), but you wouldn't happen
to have the complete fix for this, would you? I tried what I thought
you were describing and it didn't work. Also, does using _tzset mean
TZ has to be set? The _tzset page on msdn.microsoft.com seem to imply
it did, but the Time Management Routines page seems to imply that TZ
only overrides the system settings. I tried both ways and neither
worked.
I found Chuck's patch on mail-archive.com, but your mail seems to imply
that Chuck's patch was unecessary and there is a smarter fix.
I did find a note on the gmtime page that says that "The target
environment should try to determine whether daylight savings time is in
effect. The C run-time library assumes the United States's rules for
implementing the calculation of Daylight Saving Time (DST)". Does
anybody know what this means?
Derek
--
Derek Price CVS Solutions Architect ( http://CVSHome.org )
mailto:[EMAIL PROTECTED] OpenAvenue ( http://OpenAvenue.com )
--
"It's difficult to work in a group when you're omnipotent."
-Q, "Deja Q"
_______________________________________________
Bug-cvs mailing list
[EMAIL PROTECTED]
http://mail.gnu.org/mailman/listinfo/bug-cvs
