"Robert Simonson" <[EMAIL PROTECTED]> writes:
> Thanks for your input Jeff. I've revised this patch some. Here it is...
I'll commit in a day or so if nobody has any concerns. Feel free to
remind me Monday if I haven't done so.
There are some tweaks which I will make, as mentioned below...
> apr/file_io/unix/filestat.c
> ===================================================================
> --- filestat.c.old Thu Mar 14 11:20:04 2002
> +++ filestat.c Thu Mar 14 13:19:02 2002
> @@ -245,3 +245,26 @@
> return apr_stat(finfo, fname, wanted | APR_FINFO_LINK, cont);
> }
>
> +APR_DECLARE(apr_status_t) apr_file_time_set(const char *fname,
> + apr_time_t atime,
> + apr_time_t mtime)
> +{
> +#if APR_HAVE_UTIME_H
It seems nicer to do a check like "#if HAVE_UTIME" (i.e., check for
the function) here. The APR_HAVE_UTIME_H info is still useful to
have. I'll tweak configure to figure out HAVE_UTIME.
> apr/configure.in
> ===================================================================
> --- configure.in.old Thu Mar 14 13:26:00 2002
> +++ configure.in Thu Mar 14 13:42:05 2002
> @@ -903,6 +904,7 @@
> AC_SUBST(signalh)
> AC_SUBST(sys_waith)
> AC_SUBST(pthreadh)
> +AC_SUBST(utime)
That should really be "AC_SUBST(utimeh)".
I'll probably commit dummy versions to win32, os2, and netware. With
that I can safely modify one of the APR test programs to try to set
the time on a test file and if it worked (not APR_ENOTIMPL) it can
then try to retrieve the time via apr_stat() and verify that it was
set correctly.
--
Jeff Trawick | [EMAIL PROTECTED]
Born in Roswell... married an alien...
